CN PTCIP Ver 4.2 Redline

Clean REDACTED Version

PTC Implementation Plan (PTCIP)

Submitted in fulfillment of FRA Regulations Part 236, Subpart I, Section 236.1011 August 24, 2012 PTCIP

PREFACE

Request for Confidential Treatment Pursuant to 49CFR§209.11

The following is Canadian National Railway Company‟s Positive Train Control Implementation Plan (“PTCIP”), submitted in fulfillment of 49CFRPart 236, Subpart I, § 236.1011.

As authorized by 49CFR§236.1009(e)(3), CN requests confidential treatment, pursuant to 49CFR§209.11, for certain portions of the document on the basis that these portions contain: (1) sensitive security information as defined in 49CFRPart 15 (“SSI”), (2) confidential trade secrets or other proprietary commercial and/or financial information that are exempt from the mandatory disclosure requirements of the Freedom of Information Act, 5 U.S.C. §552, and/or required to be held in confidence under the Trade Secrets Act, 18 U.S.C. §1905, and (3) safety analysis records protected from public disclosure under 49 U.S.C. §20118.

Information redacted as confidential includes tables, figures and narrative relating to (1) the installation risk analysis, such as risk factor levels, risk factor weights, and risk rankings; (2) line segment characteristics, including freight, passenger, and TIH/PIH volumes, and the track attributes; and (3) the sequence and schedule for deployment of the PTC system. These portions of CN‟s PTCIP contain sensitive security information the public disclosure of which would be detrimental to transportation safety and security. This information is also integral to the analysis of safety risks CN conducted in order to implement the PTC system in a manner that addresses areas of greater risk before areas of lesser risk, as required by 49 U.S.C. §20157(a)(2). Finally, specific information regarding CN‟s routing of certain traffic, its operations, and the attributes of particular subdivisions and line segments constitutes confidential business information that, if publicly disseminated, could result in competitive harm.

In accordance with the requirements of 49CFRPart 15, CN has properly marked every page of the document and its appendices (including pages that do not contain SSI) to indicate that the document contains “SENSITIVE SECURITY INFORMATION.” In addition, the document has been marked with the statement “CONTAINS CONFIDENTIAL INFORMATION,” as required by 49CFR§209.11(d).

CN is also submitting a redacted “public” version of the PTCIP in which all sensitive or confidential information has been removed. Because this version does not contain SSI or confidential material, it does not include the markings required by 49CFR§209.11(d) and 49CFRPart 15. Finally, as specified by 49CFR§236.1009(e)(3), to assist FRA in efficiently and correctly reviewing requests for confidentiality, CN is also submitting a version of its PTCIP which highlights the portions of the document that have been redacted from the public version. Because this version of the document contains SSI and confidential material, it has been properly marked with the designations referenced above.

CN requests that only the Redacted Version of the PTCIP be placed on the public docket or otherwise disclosed.

1 PTCIP

Date Revision Description Author 16Apr2010 3.0 Original PTCIP submission document – Conditionally Sal Macri, approved by FRA. Dwight Tays 12Aug2010 4.2 Document updated per FRA requests in conditional Sal Macri, acceptance letter. Document updates include: Dwight Tays 1. Include Letters of Understanding between CN and other Railways – Sections 5.1, 9.4 and Appendix D 2. Clarify CN intent with respect to interoperability with Tenant Railways – Section 5.4 3. Clarify EJ&E traffic volume data – Section 1.1.4 and 6.1.1 4. Clarify Class II/III tenant roads that operate more than 20 miles on CN tracks – Section 5.2.3 5. Change request for exception for Sprague Subdivision to notice of pending waive rrequest – Sections 1.1.5, 8.11 (new section) and section 15 6. Clarify DeMinimis text – Section 8.10 18Apr2012 4.3 Document Updated per: Sal Macri, 1. Approved Request for Amendment – Dwight Tays Eldorado Subdivision [11] - Removal of the requirement to install a PTC system on 29 miles of the Eldorado Subdivision because less than 5 MGT of traffic has been transported over this segment for two consecutive years, no passenger traffic operates over this segment, and the segment no longer qualifies as a main line track for which a PTC system must be installed. 2. Approved Request for Amendment – CN PTC Pilot territory [12] - Change of segments that comprise the Pilot deployment group because the line segments initially identified are candidates for removal from the PTC system installation requirements after forthcoming rulemaking makes changes to the parameters of the de minimis exception 3. Approved Request for Amendment – Rolling stock Schedule and Annual PTC Goals [13] - Approval to (1) amend the scheduled installation of PTC equipment on locomotives (due to delays in the availability of onboard PTC equipment), and (2) include annual goals for PTC-equipped train operations (which were inadvertently omitted from the 2010

2 PTCIP

PTCIP). 4. Approved Request for Amendment –Wayside Deployment Schedule [14] - Approval to adjust the scheduled installation of PTC wayside equipment due to delays in the availability of necessary equipment and software. 5. V-ETMS system name updated to I-ETMS throughout. 6. The FRA has issued a Type Approval FRA- TA-2011-02 for the I-ETMS PTCDP (version 2). This document has been updated to replace references to an anticipated type approval for the I-ETMS system to the type approved I-ETMS system. 7. References to the PTCDP in Section 3 of this PTCIP have been updated to match the I- ETMS PTCDP Rev. 2 dated June 1, 2011. References to Federal regulations have been formatted consistently throughout the document. 24 Aug 12 4.4 Revised per FRA and internal CN comments based on Sal Macri, approved RFAs (included in Appendix) Dwight Tays

3 PTCIP

TABLE OF CONTENTS PAGE 1. Introduction ...... 10 1.1. PTC Overview ...... 12 1.1.1. PTC Deployment ...... 12 1.1.2. Provision of Statutory Functionality ...... 13 1.1.3. PTC System Definition ...... 13 1.1.4. EJ&E Acquisition ...... 13 1.1.5. Main Line Track Segments ...... 14 1.1.6. Installation Risk Prioritization Methodology ...... 14 1.1.7. Organizational Relationships ...... 15 1.1.8. Request for Amendment of a PTCIP [§236.1009(a)(2)(ii)] ...... 17 1.2. Goals and Objectives ...... 19 1.2.1. Quality and Safety ...... 19 1.2.2. System Coverage ...... 20 1.2.3. Progressive Risk-Based Deployment ...... 20 1.2.4. Interoperability ...... 20 1.2.5. Regulatory Compliance ...... 20 1.3. Success Criteria ...... 21 1.3.1. Long Term Goal Metrics ...... 21 1.3.2. Intermediate Goal Metrics ...... 21 1.4. Applicability ...... 23 1.5. Document Overview ...... 23 1.6. Acronyms and Definitions ...... 25 2. Applicable Documents ...... 28 3. Functional Requirements [§236.1011(a)] ...... 29 3.1. I-ETMS Development Plan Overview ...... 29 3.2. I-ETMS Technical Description ...... 30 3.3. I-ETMS Functional Description ...... 33 3.3.1. I-ETMS Components ...... 33 3.3.2. I-ETMS System Description ...... 33 3.3.3. I-ETMS Safety Architecture ...... 34 3.3.4. I-ETMS Functional Requirements ...... 34 4. Compliance [§236.1011(A)(2)] ...... 38 4.1. PTC System Certification ...... 38 4.1.1. Utilization of Existing Type Approval and PTCDP ...... 38 4.1.2. Certifying the Validity of Type Approval ...... 38 4.1.3. Handling of Unique Aspects of the PTCDP and Type Approval ...... 38 4.1.4. Deliverables ...... 39 4.2. Risk Assessment ...... 40 4.2.1. Performance Risks ...... 40 4.2.2. Deployment Risks ...... 41 4.2.3. Compliance Risks ...... 42 4.2.4. Technical Risks ...... 44 5. Interoperability [§236.1011(a)(3)] ...... 45 5.1. Railroad Agreement Provisions Relevant to Interoperability [§236.1011(a)(3)(i)] ...... 45 5.2. Types of Interoperability ...... 46 5.2.1. Native Interoperability ...... 46 5.2.2. Onboard Functional Interoperability ...... 47 5.2.3. Unequipped Operation ...... 47 5.3. Technology Applicable to Interoperability [§236.1011(a)(3)(ii)]...... 47

4 PTCIP

5.3.1. Technical interoperability ...... 48 5.3.2. Semantic Interoperability ...... 48 5.3.3. Organizational interoperability ...... 48 5.4. Obstacles to Interoperability [§236.1011(a)(3)(iii)] ...... 48 6. Designating Track as Main Line or Non-Main Line [§236.1011(a)(8)] ...... 50 6.1. CN Network Descriptions ...... 50 6.1.1. EJ&E Acquisition ...... 50 6.1.2. P&I Railroad ...... 50 Once final information is available, if there is a need to adjust the PTC implementation plan on the P&I Railroad, this will be requested by a RFA...... 51 6.1.3. CN Network Changes – 2008 to 2010 ...... 51 6.2. CN Main Line Track Segments ...... 51 6.2.1. CN Subdivisions Exceeding 5 MGT in 2008 ...... 53 6.2.2. Subdivision Segments Exceeding 5 MGT ...... 54 6.2.3. Subdivisions with Regularly Scheduled Passenger Trains ...... 55 6.2.4. Restricted Speed Track Revisions to Line Segment Mileages ...... 55 6.2.5. Final CN Main Line Track Segment Mileages...... 56 6.3. Summary of Technical Notes on CN Data ...... 57 6.4. Foreign Owned Line Segments ...... 58 6.5. MTEA Requests ...... 59 7. Installation Risk Analysis [§236.1011(a)(4)] ...... 61 7.1. The Rail Network ...... 61 7.2. Risk Factors, Risk Factor Levels, and Risk Factor Weights ...... 62 7.2.1. Risk Factor 1: Annual Million Gross Ton (MGT) ...... 63 7.2.2. Risk Factor 2: Presence and Volume of Passenger Traffic ...... 63 7.2.3. Risk Factor 3: Presence and Volume of TIH/PIH Material (Loads and Residue) Transported ...... 64 7.2.4. Risk Factor 4: Number of Tracks ...... 65 7.2.5. Risk Factor 5: Method of Operation ...... 66 7.2.6. Risk Factor 6: Speed of Train Operations ...... 67 7.2.7. Risk Factor 7: Track Grades ...... 68 7.2.8. Risk Factor 8: Track Curvature ...... 68 7.3. Overall Risk Ranking ...... 70 8. Deployment Sequence and Schedule [§236.1011(a)(5)] ...... 72 8.1. CN Key Service Corridors ...... 72 8.2. CN PTC Corridor Deployment Approach...... 72 8.3. CN Deployment Groupings ...... 73 8.3.1. Pilot Deployment Group: [§236.1011 (a)(4)&(5)] ...... 73 8.3.2. Central Deployment Group: Chicago to Memphis ...... 73 8.3.3. Gulf Deployment Group: Memphis to New Orleans ...... 74 8.3.4. East Deployment Group: Chicago to Port Huron ...... 74 8.3.5. North Deployment Group: Chicago to Ranier...... 75 8.4. Deployment Group Weighted Risk Ranking [§236.1011(a)(5)(iii)] ...... 76 8.5. Deployment Group Traffic Characteristics [§236.1011(a)(5)(i)] ...... 77 8.6. Deployment Group Operational Characteristics [§236.1011(a)(5)(ii)] ...... 77 8.7. Deployment Group Attributes [§236.1011(a)(5)(iii)] ...... 78 8.7.1. Grade, Curvature, Switches & Road Crossings ...... 78 8.7.2. Rail to Rail Crossings at Grade ...... 78 8.7.3. Movable Bridges ...... 79 8.7.4. Passenger Operations ...... 79 Presence of Other Traffic – Shared Routes ...... 80

5 PTCIP

8.8. Proposed Deployment Schedule ...... 82 Exceptions to Risk Based Prioritization [§236.1011 (a)(9)] ...... 84 8.8.1. Corridor Deployment ...... 84 8.8.2. Geographically Proximate Subdivisions ...... 84 8.8.3. Subdivisions with Limited Segments of Passenger Operations ...... 84 8.9. De-Minimis Exception Requests [§236.1005 (b)(4)(ii)] ...... 85 8.10. Sprague Subdivision – Description and Overview ...... 85 9. Rolling Stock [§236.1011(a)(6)] ...... 89 9.1. CN Locomotive Fleet Overview ...... 89 9.2. Locomotives to be PTC Equipped [§236.1011(a)(6)(i)] ...... 89 9.3. Rolling Stock PTC Implementation Schedule [§236.1011(a)(6)(ii)] ...... 90 9.3.1. Establishment of Annual Goals for PTC-Equipped Train Operations ...... 92 9.4. Tenant Railroads [§236.1011(a)(6)(iv)(A) and (B)] ...... 93 10. Wayside Devices [§236.1011(a)(7)] ...... 95 10.1. Wayside Device Equipment ...... 95 10.1.1. Wayside Interface Units ...... 95 10.1.2. Wayside/Base Communications Packages ...... 96 10.2. Wayside Device Tabulations ...... 96 10.3. Wayside Deployment Schedule [§236.1011(a)(5)&(7)] ...... 97 11. Submittal Dates for PTCDP and PTCSP [§236.1011(a)(10)] ...... 99 12. Strategy for Full PTC System Deployment [§236.1011(b)] ...... 100 13. Main Line Track Exclusion Addendum [§236.1019] ...... 101 13.1. MTEA General ...... 101 13.2. MTEA Request – Freeport/Chicago Subdivision and St. Charles Airline ...... 102 13.2.1. Freeport Subdivision ...... 102 13.2.2. Chicago Subdivision ...... 103 13.2.3. MTEA Request ...... 103 13.3. MTEA Request – Memphis Subdivision MP 380.4 to MP 394.3 ...... 104 13.4. MTEA Request – Y&MV Main ...... 106 13.5. MTEA Request – Whirlpool Bridge ...... 107 13.6. MTEA Request – McComb Subdivision MP 904.4 to MP 908.6 ...... 109 13.7. MTEA Request – Rouse’s Point Sub MP 1.18 to 0.0 ...... 109 14. De Minimis Track Exclusion Requests [§236.1005] ...... 112 14.1. De Minimis General ...... 112 14.2. De Minimis Request – Cherokee Subdivision ...... 114 14.3. De Minimis Request – Minneapolis Subdivision...... 117 Appendix A: Line Segment Attributes Detailed Tables ...... 120 Appendix B: Risk Factor Prioritization Model ...... 129 1. Introduction ...... 131 2. Risk Prioritization Model Approach ...... 132 2.1. Identification of Risk Factors ...... 132 2.2. Estimation of Risk Factor Weights ...... 134 2.2.1. Review of Previous Applicable Studies and FRA Data ...... 136 2.3. Definition of Risk Factor Levels ...... 139 2.4. Assignment of Risk Factor Levels to Subdivisions ...... 139 3. Description of Risk Factors and Quantification of Risk Factor Levels and Weights ...... 140 3.1. Risk Factor #1: Annual Million Gross Ton (MGT) Level ...... 140 3.1.1. Risk Factor Overview ...... 140 3.1.2. Quantification of Risk Factor Weight ...... 141 3.1.3. Quantification of Risk Factor Levels ...... 141 3.2. Risk Factor #2: Presence and Volume of Passenger Traffic ...... 142

6 PTCIP

3.2.1. Risk Factor Overview ...... 142 3.2.2. Quantification of Risk Factor Weight ...... 143 3.2.3. Quantification of Risk Factor Levels ...... 143 3.3. Risk Factor #3: Presence and Volume of Toxic Inhalation Hazard / Poison Inhalation Hazard (TIH/PIH) Material (Loads and Residue) Transported ...... 144 3.3.1. Risk Factor Overview ...... 144 3.3.2. Quantification of Risk Factor Weight ...... 144 3.3.3. Quantification of Risk Factor Levels ...... 145 3.4. Risk Factor #4: Number of Tracks ...... 147 3.4.1. Risk Factor Overview ...... 147 3.4.2. Quantification of Risk Factor Weight ...... 148 3.4.3. Quantification of Risk Factor Levels ...... 148 3.5. Risk Factor #5: Method of Operation ...... 149 3.5.1. Risk Factor Overview ...... 149 3.5.2. Quantification of Risk Factor Weight ...... 152 3.5.3. Quantification of Risk Factor Levels ...... 152 3.6. Risk Factor #6: Speed of Train Operations ...... 153 3.6.1. Risk Factor Overview ...... 153 3.6.2. Quantification of Risk Factor Weight ...... 154 3.6.3. Quantification of Risk Factor Levels ...... 154 3.7. Risk Factor #7: Grade ...... 155 3.7.1. Risk Factor Overview ...... 155 3.7.2. Quantification of Risk Factor Weight ...... 155 3.7.3. Quantification of Risk Factor Levels ...... 155 3.8. Risk Factor#8: Curvature ...... 156 3.8.1. Risk Factor Overview ...... 156 3.8.2. Quantification of Risk Factor Weight ...... 156 3.8.3. Quantification of Risk Factor Levels ...... 157 3.9. Other Risk Factors Not Included in the Risk Prioritization Model ...... 157 4. Model Calculation Tool ...... 160 5. Risk Prioritization Model Results ...... 171 6. External References...... 172 Appendix C: Review of Previous Applicable Studies ...... 174 Appendix D: Interoperability Letters of Understanding ...... 186

List of Tables

TABLE 1 ACRONYMS ...... 25 TABLE 2 DEFINITION OF TERMS...... 26 TABLE 3 PERFORMANCE RISK 1 ...... 40 TABLE 4 PERFORMANCE RISK 2 ...... 41 TABLE 5 DEPLOYMENT RISK 1 ...... 41 TABLE 6 DEPLOYMENT RISK 2 ...... 42 TABLE 7 COMPLIANCE RISK 1 ...... 42 TABLE 8 TECHNICAL RISK 1 ...... 44 TABLE 9 CN SUBDIVISIONS WITH TRAFFIC VOLUMES OVER 5 MGT IN 2008 ...... 53 TABLE 10 LINE SEGMENTS WITH REGULARLY SCHEDULED PASSENGER TRAINS ...... 55 TABLE 11 LINE SEGMENTS WITH RESTRICTED SPEED TRACK ...... 55 TABLE 12 CN MAIN LINE TRACK SEGMENTS ...... 56

7 PTCIP

TABLE 13 CN MAIN LINE TRACK SEGMENTS – EXCLUDING MTEA TRACKS ...... 61 TABLE 14 ANNUAL MGT RISK FACTOR LEVELS ...... 63 TABLE 15 DAILY PASSENGER TRAIN RISK FACTOR LEVELS ...... 64 TABLE 16 ANNUAL CAR VOLUME OF TIH/PIH RISK FACTOR LEVELS ...... 65 TABLE 17 NUMBER OF TRACKS RISK FACTOR LEVELS ...... 66 TABLE 18 METHODS OF OPERATION RISK FACTOR LEVELS ...... 67 TABLE 19 TRAIN SPEED RISK FACTOR LEVELS ...... 68 TABLE 20 TRACK GRADE RISK FACTOR LEVELS...... 68 TABLE 21 TRACK CURVATURE RISK FACTOR LEVELS ...... 69 TABLE 22 RISK FACTOR WEIGHTS ...... 70 TABLE 23 RISK FACTOR PRIORITY RANKING...... 70 TABLE 24 LINE SEGMENT RISK RANKING ...... 71 TABLE 25 PILOT DEPLOYMENT GROUP ...... 73 TABLE 26 CENTRAL DEPLOYMENT GROUP ...... 73 TABLE 27 GULF DEPLOYMENT GROUP ...... 74 TABLE 28 EAST DEPLOYMENT GROUP ...... 75 TABLE 29 NORTH DEPLOYMENT GROUP ...... 75 TABLE 30 DEPLOYMENT GROUP TRAFFIC CHARACTERISTICS ...... 77 TABLE 31 DEPLOYMENT GROUP OPERATIONAL CHARACTERISTICS ...... 77 TABLE 32 GRADE, CURVATURE, SWITCHES & ROAD CROSSINGS BY DEPLOYMENT GROUP ...... 78 TABLE 33 RAIL TO RAIL CROSSINGS AT GRADE BY DEPLOYMENT GROUP ...... 78 TABLE 34 MOVABLE BRIDGES BY DEPLOYMENT GROUP ...... 79 TABLE 35 ANNUAL PASSENGER TRAIN OPERATIONS BY DEPLOYMENT GROUP...... 79 TABLE 36 PASSENGER STATIONS BY PTC DEPLOYMENT GROUP ...... 80 TABLE 37 SHORTLINE TRAFFIC BY DEPLOYMENT GROUP ...... 80 TABLE 38 CN LOCOMOTIVE FLEET ...... 89 TABLE 39 PTC EQUIPPED LOCOMOTIVES ...... 89 TABLE 40 PTC ONBOARD INSTALLATION SCHEDULE AND % COMPLETION ...... 91 TABLE 41 PROPOSED GOALS FOR PTC-EQUIPPED TRAIN OPERATIONS ...... 93 TABLE 42 WIU INSTALLATIONS ...... 95 TABLE 43 WAYSIDE DEVICE TABULATIONS ...... 96 TABLE 44 PERCENTAGE OF WIUS AND BCPS INSTALLED ...... 98 TABLE 45 TRAFFIC CHARACTERISTICS BY DEPLOYMENT GROUP ...... 121 TABLE 46 OPERATING CHARACTERISTICS BY DEPLOYMENT GROUP ...... 122 TABLE 47 TRACK ATTRIBUTES TABLE ...... 123 TABLE 48 DEPLOYMENT GROUP ATTRIBUTES – RAILWAY CROSSINGS ...... 123 TABLE 49 PASSENGER TRAIN OPERATIONS ...... 126 TABLE 50 METRA PASSENGER TRAIN SUMMARY ...... 126 TABLE 51 AMTRAK PASSENGER TRAIN SUMMARY ...... 127 TABLE 4-1. RISK FACTOR WEIGHTING ...... 160 TABLE 4-2 RISK FACTOR RANGES ...... 161 TABLE 4-3 RISK PRIORITIZATION MODEL ...... 161 TABLE 4-4: RISK FACTOR: ANNUAL MILLION GROSS TON (MGT) ...... 163 TABLE 4-5: RISK FACTOR: PRESENCE AND VOLUME OF PASSENGER TRAFFIC ...... 164 TABLE 4-6: RISK FACTOR: PRESENCE AND VOLUME OF TIH/PIH MATERIAL (LOADS AND RESIDUE) TRANSPORTED ...... 164 TABLE 4-7: RISK FACTOR: NUMBER OF TRACKS ...... 165 TABLE 4-8: RISK FACTOR: METHOD OF OPERATION ...... 167 TABLE 4-9: RISK FACTOR: SPEED OF TRAIN OPERATIONS ...... 168 TABLE 4-10: RISK FACTOR: TRACK GRADES ...... 169 TABLE 4-11: RISK FACTOR: TRACK CURVATURES ...... 169

8 PTCIP

List of Figures

FIGURE 1 CN NETWORK MAP ...... 11 FIGURE 2 PTC DEPLOYMENT GROUPINGS ...... 13 FIGURE 3 ORGANIZATION CHART ...... 17 FIGURE 4 I-ETMS SYSTEM COMPONENTS ...... 31 FIGURE 5 I-ETMS SYSTEM DATA FLOW...... 33 FIGURE 6 CN PTC DEPLOYMENT SCHEDULE ...... 83 FIGURE 7 SPRAGUE SUBDIVISION ...... 88 FIGURE 10 MTEA REQUEST – WHIRLPOOL BRIDGE ...... 108 FIGURE 11 MTEA REQUEST – MCCOMB SUB MP904.4 TO MP908.6 ...... 109 FIGURE 12 MTEA REQUEST – ROUSE’S POINT SUB MP 1.18 TO 0.0 ...... 111 FIGURE 13 DE MINIMIS REQUEST – CHEROKEE SUBDIVISION ...... 116

9 PTCIP

1. Introduction CN (Canadian National Railway Company and its operating railway subsidiaries), operates the largest rail network in Canada and the only transcontinental network in North America with approximately 20,421 route-miles of track. CN is a leader in the rail industry linking customers to all three NAFTA nations with a network that spans Canada, from Halifax in the east to Vancouver and Prince Rupert in the west; and runs through the heart of mid-America, from northern Minnesota to New Orleans via Chicago and Memphis. It is the only rail network on the North American continent to connect three coasts – the Pacific, the Atlantic, and the Gulf of Mexico.

CN‟s freight revenues are derived from seven commodity groups representing a diversified and balanced portfolio of goods transported between a wide range of origins and destinations. This product and geographic diversity better positions the Company to face economic fluctuations and enhances its potential for growth opportunities. In 2008, no individual commodity group accounted for more than 19 per cent of revenues. From a geographic standpoint, CN is equally well diversified. In 2008, approximately 31 per cent of freight revenues came from transborder traffic, 26 per cent from offshore traffic, 24 per cent from Canadian domestic traffic, and 19 per cent from U.S. domestic traffic.

Approximately 85 per cent of the traffic volumes handled by CN are originated along its network. This enables the Company to capitalize on service advantages and build on opportunities to efficiently use assets.

The primary focus at CN is to run a safe and efficient railroad. While remaining at the forefront of the rail industry, CN‟s goal is to be internationally regarded as one of the best-performing transportation companies. The company‟s business strategy is guided by five core principles: providing good service, controlling costs, focusing on asset utilization, committing to safety, and developing people.

CN‟s commitment is to create value for its customers by providing quality and cost-effective service; and for its shareholders by striving for sustainable financial performance through profitable growth, solid free cash flow and a high return on investment.

CN continues to invest in various strategic initiatives to expand the scope of its business. A key initiative is the recent acquisition of a major portion of the EJ&E, which will drive new efficiencies and operating improvements on CN‟s network as a result of streamlined rail operations and reduced congestion.

The map below illustrates the CN network in the United States.

PTCIP

Figure 1 CN Network Map

PTCIP

1.1. PTC Overview This document provides an overview of CN‟s plan for implementation of Positive Train Control (PTC) in accordance with the mandate of the Railway Safety Improvement Act of 2008 (RSIA) and the requirements of the final rule published at 49CFRPart 236, Subpart I. The sections that follow this Overview address the following topics in greater detail:

a) how, where, and in what sequential order the PTC system will be deployed; b) how the PTC system provides the statutory functionality; c) whether the PTC system is defined for safety as non-vital, vital, stand-alone, or mixed under Part 236 criteria; d) identification of all main line track segments, including the method of operation, the maximum authorized speed(s), route characteristics, and signal systems for each, and any MTEAs or RFAs; e) the installation risk prioritization methodology used; and, f) all exceptions to the established deployment and risk methodologies.

1.1.1. PTC Deployment In compliance with the regulatory requirements defined in 49CFR§236.1011(a)(4), CN is deploying PTC in areas of greater risk to the public and railroad employees before areas of lower risk. The risk assessment factors and weighting criteria used to rank and prioritize line segments are discussed in more detail in sections 7 of the PTCIP and Appendix B attached. The established risk ranking methodology was used with the risk factors as required in 49 CFR 236, Subpart I, §236.1011(a)(5), to establish risk ratings for each CN subdivision where PTC is required.

CN is deploying PTC using a corridor based approach to minimize potential service impacts, maximize the efficiency of installation resources and optimize the utilization of PTC equipped locomotives. Installation and testing of the PTC system in a coordinated corridor oriented manner will help ensure safe operation as well as interoperability, and provide the information necessary to submit a PTC Safety Plan (PTCSP) as defined in 49CFR§236.1015.

The aggregated summary risk ranking for all subdivisions within PTC deployment groupings of subdivisions was tabulated and used to determine deployment group priorities. The subdivision grouping with the highest priority is targeted for PTC deployment first. Scheduling of successive deployment groupings of subdivisions is based on the aggregate risk ranking as well as evaluation of other factors such as maximizing deployment efficiency, optimizing utilization of PTC equipped locomotives and minimizing potential service disruptions.

The map below depicts the 5 proposed PTC deployment groupings.

PTCIP

Figure 2 PTC Deployment Groupings

1.1.2. Provision of Statutory Functionality The Wabtec Railway Electronics (WRE) Interoperable Electronic Train Management System (I-ETMS) being implemented by CN is a locomotive-centric train control system that uses a combination of locomotive, office and wayside data integrated via a radio network and provides functionality which satisfies the requirements of the RSIA . Specifically the I- ETMS system provides the ability to:

 Alert train crews to pending authority and speed limit violations, including passing a signal at Stop  Stop trains prior to exceeding authority and speed limits, including signals at Stop  Interrogate wayside signals, switches and broken rail detection circuits in a train route when operating in I-ETMS territory  Protect work zone limits by enforcing compliance with work zone restrictions

In addition to the functionality described above, the I-ETMS system is designed to support different railroads and their individual methods of operations. The system is designed for implementation across a broad spectrum of railroads without modification. This design approach supports interoperability across railroads as I-ETMS equipped locomotives apply consistent warning and enforcement rules regardless of track ownership. Design and development of I-ETMS has been supported by CSX Transportation Inc. (CSXT), Norfolk Southern Railway Company (NS), and Union Pacific Railroad (UPRR), as well as CN and other Class 1 railroads through the Interoperable Train Control (ITC) industry effort.

1.1.3. PTC System Definition The PTC system being deployed by CN is the Wabtec Railway Electronics‟ I-ETMS system, a vital overlay system as defined in 49CFR§236.1015(e)(2). I-ETMS is based on the Electronic Train Management System (ETMS) developed by WRE which has been approved by FRA under 49CFR§236, Subpart H for use in revenue service on BNSF Railway (FRA- 2006-23687-21), subject to certain conditions.

Additional details on the I-ETMS PTC system being deployed by CN are included in sections 3, 4 and 5 of this PTCIP as well as the Type Approved PTCDP.

1.1.4. EJ&E Acquisition On February 1st, 2009, CN completed its acquisition of the principal lines of the Elgin, Joliet & Eastern Railway Company (EJ&E). As part of the PTC planning process, CN has included the acquired EJ&E assets and has applied the same PTC evaluation process to the acquired assets that has been implemented on all other CN tracks. Due to the date of the transaction and subsequent traffic re-routing, it has been determined that use of 2008 traffic volumes for Million Gross Tons (MGT) would be less representative of expected traffic volumes under CN operations than using 2009 traffic volumes pro-rated for a full 12 month period for the acquired EJ&E subdivisions (Matteson, Leithton, Lakefront, Illinois River).

PTCIP

Therefore, we have used the 2009 traffic data for these segments. Toxic Inhalation Hazard/Poisonous by Inhalation Hazard (TIH/PIH) traffic volumes included in the PTCIP are based on actual 2008 data and are slightly greater than 2009 volumes.

1.1.5. Main Line Track Segments The CN US network includes 82 subdivisions of track that were reviewed to determine if they qualified as main line track segments under the RSIA and 49CFR§236 PTC regulations. These subdivisions include all of CN‟s US operating network with the exception of tracks where all trains are limited to restricted speed within a yard or terminal area or on auxiliary or industry tracks.

Each of the 82 CN subdivisions were evaluated according to the main line track definitions in 49CFR§236.1003 and §236.1005(b)(1)(i and ii). Using the 5MGT and regularly scheduled commuter or inter-city passenger train criteria, there are 48 CN subdivisions that qualify as main line track segments. A discussion of this evaluation is included in Section 6 of this PTCIP.

Six of the 48 subdivisions that meet the main line track criteria, had no passenger train or TIH/PIH traffic (loads or residue) in 2008 and have therefore been eliminated from the CN PTC Implementation planning and weighted risk assessment process (see Section 6 for more details). One additional line segment handles exclusively passenger traffic at restricted speed (Y&MV Main) and an MTEA request is included in section 6 for this segment of track. This leaves a total of 41 CN subdivisions and sections of subdivisions that are considered main line track requiring PTC and are discussed in further detail in subsequent sections of this PTC Implementation Plan.

One of CN's main line track segments that requires PTC installation is unique in that it is a 43.4 mile section of CN's Sprague Subdivision that dips into the US as it travels around the southern edge of Lake of the Woods. The Sprague Subdivision starts at Rainy River, Ontario, Canada (MP 0.0) and terminates at Winnipeg, Manitoba, Canada (MP 145.2) but traverses a short section of Minnesota from MP 1.6 to MP 45.0. Implementation of PTC on this portion of track will create significant operational hardship and expense due to its unique operating characteristics. CN has fully included this subdivision in this PTCIP and the CN PTC Deployment Plan, but intends to seek a separate waiver or exemption from PTC installation for this subdivision under general FRA regulations (i.e., outside the PTCIP process). Full details of the operating issues and potential risk mitigation alternatives will be included in the waiver request but a short summary of the specific issues is included in this PTCIP in Section 8.11.

1.1.6. Installation Risk Prioritization Methodology The risk prioritization model used by CN is a basic weighted score approach in which a number of risk factors were assigned integer scores corresponding with level of risk ranging from 0 (lowest risk) up to 5 (highest risk) for each of the CN subdivisions to be equipped with PTC. Each risk factor was also assigned a weight which provided an indication of the “relative importance” of the factor in determining the overall risk ranking. Equation 1 below shows how, for n risk factors, a relative risk score was generated for each subdivision by

PTCIP

multiplying the integer score assigned to the subdivision for a given factor (FRi) by the weight assigned to that factor (FWi), and summing the products of the n risk factors.

(Equation 1) Relative Risk Score for Subdivision =

A summary of the risk prioritization model is provided in Section 7, and additional model details are provided in the Risk Prioritization Model report included in Appendix B.

1.1.7. Organizational Relationships

The PTC implementation team was organized to provide the highest level of Executive support and skilled, experienced leaders in every technical area of the project. Each level of the project team has clear roles and responsibilities and access to a wealth of knowledgeable resources within the organization.

1.1.7.1. Steering Committee The Steering Committee‟s role is critical for the success of the project. Composed of CN executives from several functions of the organization, together they will provide guidance, contribute valuable input on implementation plans and roll-outs, and help resolve issues and remove any road blocks. They represent the Stakeholders and Sponsors and as such, will approve budgets and final deliverables.

- James S. Bright, VP and Chief Information Officer - Keith Creel, Executive VP and Chief Operating Officer - Sameh Fahmy, Senior VP Supply Management, Engineering and Mechanical - Ghislain Houle, Vice-President Financial Planning - Paul Miller, Chief Officer Safety and Transportation - Jim Vena, Senior VP – Southern Region

1.1.7.2. Program Manager Accountable and responsible for the end-to-end delivery of PTC, the Program manager will work to establish business requirements, roadmap, timelines, deliverables and budgets. He will also assess the need for outside help and oversee contract negotiations. As the link between the Steering Committee and the Leadership Team, he will also provide guidance and approve deliverables.

1.1.7.3. Project Manager Using a Project Management Institute (PMI) inspired methodology; the Project manager will oversee the assessment and planning phase for the submission of the PTCIP, execution of the plan, testing and implementation of all of the PTC components. He will work closely with the Program Manager to define and manage scope, high level schedule and resource plan. He is responsible for tracking of deliverables and budgets, for coordinating all project activities and for providing relevant status information to the team.

PTCIP

1.1.7.4. Senior Managers (by areas) The PTC system has been divided into 4 technical areas: the back office application, the communication network, the wayside systems and the locomotive on-board systems. Senior managers have been assigned to each component and are responsible for the gathering of data, design, sourcing, delivery and quality control of their respective area. In order to do so, they have access to internal resources and strategic partners with required skill sets. In addition, CN‟s Supply Management department will support the acquisition processes (RFI, RFP, RFQ) and contract negotiation during the PTC project.

1.1.7.5. Strategic Partners CN has elected to work with strategic partners with proven track records in the railway and safety industry. Their deliverables are part of the overall plan and specifically aligned with their expertise and experience.

Wabtec Railway Electronics (21200 Dorsey Mill Road, Germantown MD 20876): WRE has been selected to provide the Interoperable Electronic Train Management System (I- ETMS) to satisfy the statutory functionality as defined in the RSIA. As such, they have provided the content of the PTCDP in compliance with §236.1013. Rail Safety Consulting, LLC (1151 Pittsford-Victor Rd. Pittsford, NY 14534): RSC is a consulting organization with detailed knowledge of safety designs and operating rules, processor-based systems and has worked with several railroads on their PTC plan. They have been contracted by CN to validate technical assumptions related to the system risk assessment and assist with the writing of the PTCIP. The mandate could be extended as project requires. Wayside Equipment Vendors: CN plans to evaluate and test WIU equipment from a number of equipment vendors to determine which equipment is best suited for use in each of the various wayside PTC applications (electronic control equipment, relay based interlockings, dark territory switches, etc). Equipment selected for use will be expected to meet accepted industry standards for vital wayside signaling equipment and CN will work closely with equipment manufacturers to ensure appropriate documentation is available to support the required PTCSP submission. CN will work closely with selected wayside equipment vendors to ensure that all equipment used for wayside PTC applications is installed and maintained in accordance with manufacturers recommendations.

1.1.7.6. Other technical resources During each phase of the project, technical resources will be made available as required to provide expertise and collaborate regarding various deliverables.

1.1.7.7. Change Management, Transition to Core & Operational Organizations Once delivered, the PTC system and its components will be integrated into the Operations and Maintenance Manual as per §236.1039. To properly plan, manage and provide training to each area and personnel of the organization, a Change Management team will work on transition to core activities to engage the right department at the right time and provide the right level of information and training, as described in §236.1041-49.

PTCIP

Figure 3 Organization Chart

1.1.8. Request for Amendment of a PTCIP [§236.1009(a)(2)(ii)]

PTCIP

This subsection describes how CN will make and file a Request For Amendment (RFA) of its PTCIP in accordance with §236.1021.

This revision of the CN PTCIP is based on the submission and approval of four (4) RFAs describing changes to the CN implementation of PTC. The revised document reflects the impact of the RFAs on the content of the PTCIP as can be seen as changes in the “redline” version as submitted. The RFAs were determined to be needed and submitted according to the process described below.

On an annual basis, CN will review operations for any routing changes, such as traffic density changes above or below the 5 Million Gross Tons threshold or the presence or absence of both TIH/PIH hazardous materials or passenger trains and other changes listed in §236.1005(b). The intent of this review will be to identify any changes made, or planned, to the system that requires an RFA to the PTCIP. When it is determined that any of the changes identified by the review, “Add, subtract, or otherwise materially modify one or more lines of railroad for which installation of a PTC system is required”, CN will prepare a Request for Amendment of this PTCIP as per §236.1021. The four (4) RFAs submitted to date are identified in the “Applicable Documents” Section 2 of this PTCIP revision.

Prior to CN submitting each RFA or changing or altering traffic patterns, they were reviewed by the CN PTC Steering Committee. The purpose of this internal review was to ensure that all requisite factors and data have been included in the internal evaluation and to update CN Senior Executives regarding the revised PTC deployment and funding requirements. For annual reporting, the internal review will be scheduled as soon as practical but shall be completed in sufficient time to allow the RFA to be submitted to the FRA in conjunction with the annual PTCIP update required by April 16 each year.

Throughout the implementation of the CN PTC system, configuration management will be performed in accordance with the CN Configuration Management Plan (CMP). The CMP establishes the configuration management practices that will implement and maintain an effective and timely method for defining and controlling the configuration of all equipment. This includes design, manufacturing and installation of the fixed facilities, car-borne and wayside equipment, and all interfaces. A software configuration management standard will be employed to properly track and control revisions to software against an established baseline software version. In addition, at the project level, operating procedures will be in place to provide for the proper updating, verification, control and installation of software throughout the project life-cycle, through and including field testing and in-service commissioning.

The configuration management of all FRA safety submittal documents (PTCIP, PTCDP, and PTCSP) is covered by this CMP. CN will review and approve the PTC vendor(s) Configuration Management Plan to ensure that it is consistent with the CN CMP.

In accordance with 49CFR§236.1039, hardware, software, and firmware revisions will be documented in the Operations and Maintenance Manual per the practices established in the CN CMP.

PTCIP

1.2. Goals and Objectives

This section describes the overall goals and objectives of CN‟s PTC implementation initiative including specific objectives in the areas of quality, safety, network coverage, risk based deployment, interoperability and regulatory compliance.

The primary goal for the deployment of PTC technologies on CN‟s US network is to enhance system safety, with particular focus on the prevention of:

 train-to-train collisions  overspeed derailments,  incursions into established work zone limits  the movement of trains through improperly-positioned switches Enhancements to system safety will be achieved as a PTC vital overlay system is progressively deployed across all portions of the CN US network for which PTC deployment is required by 49CFR§236.1005(b), with all required portions of the CN US network to be fully equipped, operational, and interoperable with all tenant railroads by December 31, 2015.

Goals and objectives relating to various aspects of PTC deployment are described in additional detail below.

1.2.1. Quality and Safety Deployment of PTC technologies will be conducted in full compliance with all applicable Federal requirements, including those specified in 49CFRPart 236 Subpart I, and an acceptable level of safety will be maintained in the development, functionality, architecture, installation, implementation, inspection, testing, operation, maintenance, repair, and modification of the PTC technologies to be deployed. To ensure that an acceptable level of safety is achieved, the methodologies and activities to be defined in the PTCSP, as required by 49CFR§236.1015, will be followed, and as a part of this, CN will ensure that all vendors from whom PTC technologies are to be acquired will have an acceptable quality assurance program for both design and manufacturing processes. The “systems” approach that will be employed by CN will also help ensure safe and reliable functionality and interaction between the wayside, on-board, and office components of the PTC system, with the communications component of the system playing a crucial role in accommodating this safe and reliable interaction. This holistic view will be necessary, as it is anticipated that products from multiple vendors will be integrated into the PTC system design.

PTCIP

1.2.2. System Coverage In complying with the requirements of §236.1005, CN will be installing PTC technologies on 41 of its 82 US subdivisions, corresponding to approximately 62% of CN‟s 6,213 total US network route miles. Of the roughly 3,720 route miles to be equipped, approximately 1,180 route miles accommodated passenger operations in 2008. A total of 190,276 cars of TIH/PIH were handled on the 3,720 route miles that will be equipped with PTC, which represents 97% of the total of 196,057 cars of TIH/PIH handled on the CN US track network. Implementing PTC on subdivisions where passenger traffic and/or a substantial amount of TIH/PIH traffic is present will reduce the risk associated with catastrophic accidents involving passenger trains and TIH/PIH materials, in keeping with Congress‟s mandate, as outlined in RSIA08.

1.2.3. Progressive Risk-Based Deployment The progressive deployment of PTC technologies across CN‟s subdivisions will take place in a manner such that, to the extent practical, the PTC system will be implemented to address areas of greater risk to the public and railroad employees before areas of lesser risk. Deployment of PTC on the CN network will focus on a corridor oriented approach where higher risk corridors between major terminals are equipped in priority order. CN will also achieve progressive implementation of onboard systems and deployment of PTC-equipped locomotives such that the safety benefits of PTC are achieved through incremental growth in the percentage of equipped controlling locomotives operating on PTC lines.

1.2.4. Interoperability The PTC system will provide for interoperability between CN and all tenant railroads, as technical, semantic, and organizational interoperability will be achieved to enhance the ability of CN and its tenants to operate together safely. Interoperability between CN and its tenants will be achieved through product testing, industry partnership, common technology, and standard implementation. CN and its tenants will work closely together throughout the PTC deployment process to ensure that all aspects of interoperability are fully addressed, and this partnership will be on-going as the railroads proceed to operate on these equipped portions of the CN network into the foreseeable future.

1.2.5. Regulatory Compliance In order to meet the December 31, 2015 deadline mandated by Congress, CN has developed this PTCIP in accordance with §236.1011 and submitted the original version by the required April 16, 2010 deadline. CN also provided a [9] in accordance with §236.1013, by the required April 16, 2010 deadline. CN has now determined that it will be following the PTC design and operation described in the Type Approval for I-ETMS (FRA-TA-2010-002) which is based on the I-ETMS PTCDP Revision 2 of June 1, 2011 [5]. This reference is a change from the previously submitted PTCDP and replaces it completely.

It is CN‟s intent tosubmit a PTCSP and achieve FRA PTC System Certification by the end of September 2013 and, following Certification, to deploy PTC on all required portions of the network by August 15, 2015, such that CN‟s PTC system will be fully operational by December 31, 2015 per regulation.

PTCIP

1.3. Success Criteria This section of the PTCIP describes the metrics that will be applied to gauge the success of long term and intermediate implementation goals. For clarification, when referred to in this section, long term goals shall refer to CN‟s implementation milestones from a system point of view. Intermediate goals shall refer to CN‟s implementation milestones from a subdivision point of view.

1.3.1. Long Term Goal Metrics To gauge long term goals, CN shall use the following metrics for System PTC Implementation and Locomotive Installation. The remaining metrics will be on a subdivision to subdivision basis and are described in Section 1.3.2 Intermediate Goal Metrics.

1.3.1.1. PTC System Implementation

A subdivision will be considered complete when PTC System Certification is received by CN as set forth in §236.1015(a). CN sets forth the following yearly metrics for the number of subdivisions it shall have that are capable of running PTC once Certified:

 2012: 2 of 41 subdivisions have been completed = 7% of track.  2013: 10 of 41 subdivisions have been completed = 25% of track.  2014: 30 of 41 subdivisions have been completed = 71% of track.  2015: 41 of 41 subdivisions have been completed = 100% of track.

1.3.1.2. Locomotive Installation

Since CN does not assign its locomotives per subdivision, it is appropriate to consider the equipping of rolling stock as a long term goal. CN sets forth the following yearly metrics for the number of locomotives that it shall have equipped with PTC:

 2010: 12 of 1000 locomotives have been equipped = 1%  2011: 34 of 1000 locomotives have been equipped = 3%  2012: 134 of 1000 locomotives have been equipped = 13%  2013 394 of 1000 locomotives have been equipped = 39%  2014: 694 of 1000 locomotives have been equipped = 69%  2015: 1000 of 1000 locomotives have been equipped = 100%

1.3.2. Intermediate Goal Metrics Intermediate goals shall refer to those milestones that can best be used on a subdivision to subdivision basis. When all of these intermediate goals have been completed, a subdivision shall be considered cutover to PTC operations.

PTCIP

1.3.2.1. Infrastructure Installation Completed

Infrastructure installation for a subdivision shall be completed when the following have been installed and tested for functionality:

 100% of the communication system  100% of the track infrastructure  100% of the waysides

1.3.2.2. GIS Validated

There are two intermediate goals on each subdivision that are a result of Geographic Information System (GIS) data. GIS data shall be considered validated for a subdivision when the following are completed:

 Track Survey Completed  Track Database Validated & Verified

1.3.2.3. Field Testing Completed

The completed field testing shall conform with §236.1015(d)(10). This testing will be made up of the following:

 Host Railroad PTC Operation Tested  Interoperable PTC Functionality Tested

1.3.2.4. Training Completed

As an intermediate goal, training shall be considered completed once the following have been accomplished:

 A sufficient number of dispatchers will have been trained to operate each subdivision that has been cut over to PTC.  A sufficient number of engineers will have been trained to operate all locomotives that are functioning under the PTC umbrella.  A sufficient number of field maintainers and supervisors will have been trained to service all PTC-equipped track that has been put into service as such.

1.3.2.5. PTCSP Submitted

As put forth in §236.1015, the host railroad is required to submit a PTCSP in order to get its subsequent PTC System Certificate. This intermediate goal shall be considered complete once the PTCSP has been submitted to the FRA.

PTCIP

1.3.2.6. PTC System Certification Received

§236.1015(a) states that the “receipt of a PTC System Certification affirms that the PTC system has been reviewed and approved by the FRA in accordance with, and meets the requirements of, this part.” Once CN receives the PTC System Certification, the subdivision shall be considered operational.

1.4. Applicability RSIA requires that all carriers providing intercity or commuter rail passenger transportation or mainline freight lines carrying at least 5 million gross tons of freight annually and carrying any amount of TIH materials within the US have a system of Positive Train Control in operation by December 31, 2015. The law also goes on to require that railroads that meet the above criteria shall submit to the Secretary of Transportation a plan for the implementation of said systems by the date required, April 16, 2010.

CN, as a carrier which meets these criteria on some of its track, will deploy PTC on those sections where it is required and provides this implementation plan in fulfillment of the statute.

Section 6 of this document contains detail by subdivision of the pertinent information required to assess the requirement for PTC deployment. Section 13 contains information on all sections of track where we are applying for an MTEA and section 14 contains information in regards to de Minimis exclusions that will not be PTC equipped.

1.5. Document Overview This section provides an overview of the organization of the PTCIP, which CN is submitting as required by 49 U.S.C. §20157 and §236.1005 prior to implementing the PTC system.

 Section 1 describes the general objectives, applicability, and scope of the document.  Section 2 lists all applicable documents that are referenced in this PTCIP.  Section 3 describes the functional requirements that the proposed system must meet as required by §236.1011(a)(1).  Section 4 describes how the CN intends to comply with §236.1009I as required by §236.1011(a)(2).  Section 5 defines how the CN will provide for interoperability between the host and all tenant railroads as required by §236.1011(a)(3).  Section 6 identifies which track segments the railroad designates as main line and non-main line track, as required by §236.1011(a)(8).  Section 7 describes how the PTC system will be implemented to address areas of greater risk to the public and CN employees before areas of lesser risk, by evaluating multiple risk factors, as required by §236.1011(a)(4).  Section 8 defines the sequence, schedule, and decision basis for the line segments to be equipped, including the risk factors by line segment, as required by §236.1011(a)(5). Section 8 also identifies and describes the CN‟s basis for determining that the risk-based prioritization in Section 6 above is not practical as required by §236.1011(a)(9).

PTCIP

 Section 9 identifies the rolling stock that will be equipped with the PTC technology, as required by §236.1011(a)(6) and defines the schedule for implementation.  Section 10 identifies the number of wayside devices required for each line segment and the schedule to complete the installations by December 31, 2015, as required by §236.1011(a)(7).  Section 11 contains the scheduled dates for PTCDP and PTCSP delivery as required by §236.1011(a)(10).  Section 12 contains the strategy for full system-wide deployment of PTC systems beyond those line segments required to be equipped under 49CFRPart 236 Subpart I, including the criteria that will be applied in identifying those additional lines as required by §236.1011(b).  Section 13 contains the Main Line Track Exclusion Addendum as defined by §236.1019.  Section 14 contains the De-Minimis exception requests as defined by §236.1005(b)(4)(ii).

PTCIP

1.6. Acronyms and Definitions This section includes definitions of all terms, abbreviations, and acronyms required to properly interpret the Implementation Plan. The following is a list of some abbreviations and acronyms that may be used in the PTCIP: Table 1 Acronyms Acronym Meaning AAR Association of American Railroads ABS Automatic Block Signal ATC Automatic Train Control ATS Automatic Train Stop BCP Base Communication Packages BNSF Burlington Northern Santa Fe Railway CAD Computer Aided Dispatch CDU Computer Display Unit C.F.R. Code of Federal Regulations CG Central Gulf Railway CMP Configuration Management Plan CN Canadian National Railway ConOps Concept of Operations CSSSB Chicago South Shore and South Bend Railroad CSXT CSX Transportation CTC Centralized Traffic Control C&J C&J Railroad Company (Mississippi Delta Railroad) EJ&E Elgin, Joliet & Eastern Railway Company ETMS Electronic Train Management System FRA Federal Railroad Administration GIS Geographic Information System GPS Global Positioning System GTM Gross Ton Miles HESR Huron and Eastern Railway HHP High Horsepower HMI Human Machine Interface I-ETMS Interoperable Electronic Train Management System IANR Iowa Northern Railway IHB Indiana Harbour Belt Railway ITC Interoperable Train Control LHP Low Horsepower MGT Million Gross Tons MSE Mississippi Export MTEA Main Line Track Exclusion Addendum NPI Notice of Product Intent NS Norfolk Southern Railway Company PIH Poison by Inhalation Hazard PMI Project Management Institute PTC Positive Train Control PTCDP Positive Train Control Development Plan PTCIP Positive Train Control Implementation Plan

PTCIP

PTCSP Positive Train Control Safety Plan QUI Quadrennial Inspections RFA Request For Amendment RFI Request for Information RFP Request for Proposal RFQ Request for Quotation RSIA Railway Safety Improvement Act STB Surface Transportation Board TCS Train Control System TWC Track Warrant Control TIH Toxic Inhalation Hazard TMC Train Management Computer TSBY Tuscola and Saginaw Bay Railway Company UPRR Union Pacific Railroad U.S.C. United States Code

WIU Wayside Interface Unit WRE Wabtec Railway Electronics WSOB Wisconsin and Southern Railway

The following is a list of definitions of terms applicable to the PTCIP:

Table 2 Definition of Terms Term Definition

Class I railroad A railroad which in the last year for which revenues were reported exceeded the threshold established under regulations of the Surface Transportation Board (49CFR part 1201.1-1 (2008)). Fail-Safe A design philosophy applied to safety-critical systems such that the results of hardware failures or the effect of software error shall either prohibit the system from assuming or maintaining an unsafe state or shall cause the system to assume a state known to be safe. (IEEE-1483) Host railroad A railroad that has effective operating control over a segment of track. Interoperability The ability of a controlling locomotive to communicate with and respond to the PTC railroad‟s positive train control system, including uninterrupted movements over property boundaries. Main line Except as excepted pursuant to §236.1019 or where all trains are limited to restricted speed, a segment or route of railroad tracks, including controlled sidings: (1) of a Class I railroad, as documented in current timetables filed by the Class I railroad with the FRA under §217.7, over which 5,000,000 or more gross tons of railroad traffic is transported annually; or (2) used for regularly scheduled intercity or commuter passenger service, as defined in 49 U.S.C. §24102, or both. Main line track The document defined by §236.1019 requesting to designate track as other than main exclusion line. addendum NPI Notice of Product Intent as further described in §236.1013. PTC Positive Train Control to meet the requirements described in §236.1005. PTCDP PTC Development Plan as further described in §236.1013. PTCIP PTC Implementation Plan as required under 49 U.S.C. §20157 and further described in

PTCIP

§236.1011. PTC railroad Each Class I railroad and each entity providing regularly scheduled intercity or commuter rail passenger transportation. PTC System Certification as required under 49 U.S.C. §20157 and further described in §§236.1009 Certification and 236.1015. PTCSP PTC Safety Plan as further described in §236.1015 Request For A request for an amendment of a plan or system made by a PTC railroad in accordance Amendment with §236.1021. Restricted speed A speed that allows stopping in half the range of vision, short of : train, engine, railroad car, men or equipment fouling the track, stop signal, derail or switch lined properly. When a train or engine is required to move at restricted speed, the crew must keep a look out for broken rail and not exceed 20 MPH. Safety-critical Safety-critical, as applied to a function, a system, or any portion thereof, means the correct performance of which is essential to safety of personnel or equipment, or both; or the incorrect performance of which could cause a hazardous condition, or allow a hazardous condition which was intended to be prevented by the function or system to exist. (236H) A term applied to a system or function, the correct performance of which is critical to safety of personnel and/or equipment; also a term applied to a system or function, the incorrect performance of which may result in an unacceptable risk of a hazard. (IEEE- 1483) Segment of Any part of the railroad where a train operates. track Tenant railroad A railroad, other than a host railroad, operating on track upon which a PTC system is required. Track segment Segment of track Vital Function A function in a safety-critical system that is required to be implemented in a fail-safe manner. Note: Vital functions are a subset of safety-critical functions. (IEEE-1483)

PTCIP

2. Applicable Documents This section provides a complete list of all the documents and other sources referenced in the PTC Implementation Plan.

[1] 49CFR236 Subpart I, “Positive Train Control Systems; Final Rule”, Docket No. FRA-2008-0132, 15 January 2010

[2] 49CFRPart 236, Subpart H, March 5, 2005.

[3] FRA‟s PTC Implementation Plan Template as published on the FRA RSAC website.

[4] FRA‟s Risk Prioritization Model for PTC System Implementation Template as published on the FRA RSAC website.

[5] Interoperable Electronic Train Management System (I-ETMS) – Positive Train Control Development Plan (PTCDP), 1 June 2011, Version 2.0

[6] 49CFR234.211, “Grade Crossing Signal System Safety”, Subpart D, “Maintenance, Inspection, and Testing Maintenance Standards”, “Security of Warning System Apparatus” – 5 December 2005

[7] 49CFR229.135, “Railroad Locomotive Safety Standards”, “Event Recorders” – 15 January 2010

[8] MIL-STD-882C, “System Safety Program Requirements” with Notice, 1 DoD, 13 March 1996.

[9] V-ETMS Positive Train Control Development Plan, 24 March 2010, Version 1.0, as submitted by CN with original PTCIP.

[10] FRA Type Approval (FRA-TA-2011-02) for the Interoperable Electronic Train Management System (I-ETMS), dated August 26, 2011.

[11] RFA 1, Request for Amendment – Eldorado Subdivision, May 12, 2011, submitted per Docket FRA-2010-0057

[12] RFA 2, Request for Amendment – CN PTC Pilot Territory, December 16, 2011, submitted per Docket FRA-2010-0057

[13] RFA 3,, Request for Amendment – Rolling Stock Schedule and Annual PTC Goals, November 21, 2011, submitted per Docket FRA-2010-0057

[14] RFA 4, , Request for Amendment – Wayside Deployment Schedule, December 16, 2011, 2011, submitted per Docket FRA-2010-0057

Note: For dated references, only the edition cited applies. For undated references, the latest edition of the reference document applies, including amendments.

PTCIP

3. Functional Requirements [§236.1011(a)] As required by 49CFR236.1011(a)(1) this section of the PTCIP provides a general description of the functional requirements that the proposed PTC system must meet as well as a brief overview of the proposed system technology and architecture.

3.1. I-ETMS Development Plan Overview

A full and comprehensive description of the I-ETMS functionality is provided in the “Interoperable Electronic Train Management System (I-ETMS) Positive Train Control Development Plan.” The PTCDP describes how I-ETMS satisfies the mandated requirements for PTC systems as outlined in §236.1005. On 01 June 2011, the PTC Development Plan prepared by Wabtec Railway Electronics, CSX Transportation, Norfolk Southern Railway, and Union Pacific Railroad was submitted to the FRA for review and approval. The PTCDP was jointly submitted for FRA Type Approval as set forth under 49CFR236.1009(b) and included documentation as required by §236.1013. On Aug 26, 2011 the FRA granted a Type Approval FRA-TA-2011-02 for I-ETMS in accordance with 49CFR236.1009(f) and (g), and 236.1013(b).

The Interoperable Electronic Train Management System Development Plan describes development of the WRE Interoperable Electronic Train Management System, an interoperable PTC system developed in compliance with requirements and standards defined through the ITC industry effort.

A summary of the key sections of the I-ETMS PTCDP document is provided below:

 PTCDP Section 3 provides a complete description of the I-ETMS system including a list of all product components and their physical relationships in the subsystem or system, as required by 49CFR236.1013(a)(1).

 PTCDP Section 4 contains a description of the various railroad categories of operation for which I-ETMS is designed to be used as required by 49CFR236.1013(a)(2).

 PTCDP Section 5 contains an operational concepts document as required by 49CFR236.1013(a)(3).

 PTCDP Section 6 describes how I-ETMS architecture satisfies safety requirements as required by 49CFR236.1013(a)(4).

 PTCDP Section 7 provides a preliminary human factors analysis as required by 49CFR236.1013(a)(5).

 PTCDP Section 8 contains an analysis of the applicability of the requirements of subparts A-G of 49CFR as required by 49CFR§236.1013(a)(6).

PTCIP

 PTCDP Section 9 describes a prioritized service restoration and mitigation plan and a description of the necessary security measures for I-ETMS as required by 49CFR236.1013(a)(7).

 PTCDP Section 10 describes the target safety levels for I-ETMS including requirements for system availability as required by 49CFR236.1013(a)(8).

 PTCDP Section 11 provides a description of how I-ETMS will enforce authorities and signal indications as required by 49CFR236.1013(a)(9) and how I-ETMS will enforce all integrated hazard detectors in accordance with §236.1005(c)(3) as required by 49CFR236.1013 (a)(11).

 PTCDP Section 12 provides a description of the deviation which may be proposed under §236.1029(c), if applicable, as required by 49CFR236.1013(a)(10).

3.2. I-ETMS Technical Description

I-ETMS is a locomotive-centric, vital train control system designed to be overlaid on existing methods of operation and provide a high level of railroad safety through enforcement of a train‟s authorized operating limits, including:

1. protection against train to train collisions, 2. derailments due to over-speed, 3. unauthorized incursion into established work zones, and 4. operation through main track switches in improper position.

The I-ETMS system is designed to support different railroads and their individual methods of operations and is intended to be capable of being implemented across a broad spectrum of railroads without modification. This design approach supports interoperability across railroads as I-ETMS equipped locomotives will apply consistent warning and enforcement rules regardless of track ownership.

The I-ETMS system consists of components physically and logically divided into four subsystems or segments: Locomotive, Office, Communications, and Wayside (see figure below).

PTCIP

Figure 4 I-ETMS System Components

Communications Segment: The Communications Segment provides connectivity between each of the other segments.

Locomotive Segment: The Locomotive Segment refers to a set of independent On-board hardware, software, and devices that interface with locomotive control equipment (e.g. air brakes, train line) and the Communication Segment aboard a locomotive. The Locomotive Segment employs a Train Management Computer (TMC). Software running on multiple processor modules is used to perform all train control functions such as determination of current position, calculation of warning and braking distances, management of limits or restrictions conveyed by verbal or electronic mandatory directive or signal indication, management of off- board communications, and communication with the Computer Display Unit (CDU).

Office Segment: The Office Segment refers to a collection of software functions that may be distributed across multiple hardware platforms. The Office Segment is responsible for delivering data provided by railroad back office systems to I-ETMS-equipped locomotives. Data provided by existing, external railroad office systems may include train activation, engine consist, summary and detailed train consists, movement authorities, temporary speed restrictions, work zones, cautionary orders, weather, and critical alert information.

Wayside Segment: The Wayside Segment consists of those signaling appliances located in the field whose status impacts I-ETMS operations, along with any wayside interface units (WIUs) used to monitor and report their status. WIU monitors the status of one or more attached wayside

PTCIP devices and includes an interface to the Communications Segment. The WIU acquires and publishes the status of attached wayside devices via the Communications Segment to the I- ETMS Locomotive and/or Office Segments.

PTCIP

3.3. I-ETMS Functional Description Descriptions of the I-ETMS system, its primary functions, the architecture of the PTC system(s) being deployed, and a high level description of the functionality of the PTC system, subsystems, and interfaces are all found in the PTCDP. The following sections provide an overview of the key functional areas as identified in 49CFR236 Subpart I.

3.3.1. I-ETMS Components Section 3 of the I-ETMS PTCDP provides a complete description of the I-ETMS system including a list of all product components and their physical relationships in the subsystem or system, as required by 49CFR236.1013(a)(1). Please reference the following subsections within Section 3 of the PTCDP:

3.1 System Overview 3.2 Office Segment 3.3 Wayside Segment 3.4 Locomotive Segment 3.5 Communications Segment 3.6 I-ETMS Interoperability Figure 5 I-ETMS System Data Flow

3.3.2. I-ETMS System Description As required by 49CFR236.1013(a)(1), Section 3 of the I-ETMS PTCDP additionally describes the components within the I-ETMS segments and their physical relationships in the I-ETMS system. Please reference the following subsections within Section 3 of the PTCDP: 3.2 Office Segment

PTCIP

3.2.1 I-ETMS Office Segment 3.2.2 Office Server Platform 3.2.3 Office Segment External Interfaces 3.3 Wayside Segment 3.3.1 WIU Architecture 3.3.2 WIU Configurations 3.3.3 Wayside Messaging System 3.4 Locomotive Segment 3.4.1 I-ETMS Train Management Computer 3.4.2 Computer Display Unit 3.4.3 Locomotive ID Module 3.4.4 GPS Receiver 3.4.5 Locomotive Event Recorder 3.4.6 Train Control Application 3.4.7 Business Applications 3.5 Communications Segment 3.5.1 Wireless Networks 3.5.2 The Messaging System

3.3.3. I-ETMS Safety Architecture Section 6 of the I-ETMS PTCDP describes how the I-ETMS safety architecture satisfies safety requirements as required by 49CFR236.1013(a)(4). Please reference the following subsections within Section 6 of the PTCDP:

6.1 Locomotive Segment 6.2 Office Segment 6.3 Wayside Segement 6.4 Communication Segment 6.5 Safety Requirements 6.6 System Safety Process 6.7 General Onboard Processing 6.8 Locomotive Segment Interface Failures 6.9 Systemic Errors 6.10 Fault Tolerant

3.3.4. I-ETMS Functional Requirements The Concept of Operations contained in Section 5 of the PTCDP is provided as required by §236.1013(a)(3). The Concept of Operations addresses each of the PTC functional requirements and provides a thorough description of the system‟s ability to meet the requirements. For purpose of this PTCIP, each requirement is addressed by providing a cross reference to the pertinent section of Section 5 of the PTCDP, as follows:

1. §236.1005 (a)(1)(i)– Reliably and functionally prevent train to train collisions including trains operating over rail to rail at grade crossings;  Section 5.6.5 Train Movement

PTCIP

 Section 5.6.5.1Movement Authority  Section 5.6.5.2 Movement Authority Provided by Mandatory Directive  Section 5.6.5.3 Wayside Signals  Section 5.6.5.4 Cab Signals  Section 5.6.5.5 Reverse Movement  Section 5.6.5.6 Switching Mode  Section 5.6.5.7 Entry to I-ETMS Territory  Section 5.6.5.8 Exit from I-ETMS Territory  Section 5.6.5.9 Yard Limits and Restricted Limits  Section 5.6.12 Warning and Enforcement  Section 5.6.12.1Reactive (Overspeed) Warning and Enforcement  Section 5.6.12.2 Predictive Warning and Enforcement  Section 5.6.12.3 Restrictive Speed Enforcement

Railroads must address rail-to-rail crossings at grade as part of the requirement that they address train-to-train collisions. In all cases where PTC equipped lines are involved, an interlocking signal arrangement developed in accordance with subparts A through G of part 236 will be in place. I-ETMS is designed to prevent train to train collisions where interlocking signals are in place as described in the I-ETMS PTCDP Sections 5.6.5.3 Wayside Signals and 5.6.5.4 Cab Signals.

The method to be used by CN for protecting non-PTC routes at rail-to-rail crossing-at-grade will be dependent on the speed and the specific field conditions of each location, availability of alternate technologies to provide positive stop enforcement, and the presence of PTC equipped locomotives operating on the non-PTC routes.

2. §236.1005(a)(1)(ii) – Reliably and functionally prevent overspeed derailments, including derailments related to railroad civil engineering speed restrictions, slow orders, and excessive speeds over switches and through turnouts;  Section 5.6.5.9 Yard Limits and Restricted Limits  Section 5.6.6 Speed Limits and Restrictions  Section 5.6.6.1 Permanent Speed Restrictions  Section 5.6.6.2 Temporary Speed Restrictions  Section 5.6.6.3 Track Warrant Speed Restrictions  Section 5.6.6.4 Consist or Lading Speed Restriction  Section 5.6.12 Warning and Enforcement  Section 5.6.12.1 Reactive (Over-speed) Warning and Enforcement  Section 5.6.12.2 Predictive Warning and Enforcement  Section 5.6.12.3 Restricted Speed Enforcement

3. §236.1005(a)(1)(iii) – Reliably and functionally prevent incursions into established work zone limits without first receiving appropriate authority and verification from the dispatcher or roadway worker in charge, as applicable and in accordance with 49CFR Part 214;  Section 5.6.7 Work Zones  Section 5.6.12 Warning and Enforcement

PTCIP

 Section 5.6.12.2 Predictive Warning and Enforcement

4. §236.1005(a)(1)(iv) – Reliably and functionally prevent the movement of a train through a main line switch in the improper position as further described in §235.1005;  Section 5.6.11 Route Integrity Protection  Section 5.6.5.2.6 Monitored Hand-Operated Switches  Section 5.6.11.1 Switches Monitored by WIU  Section 5.6.12 Warning and Enforcement  Section 5.6.12.2 Predictive Warning and Enforcement

5. §236.1005(a)(2) - Include safety-critical integration of all authorities and indications of a wayside or cab signal system, or other similar appliance, method, device, or system of equivalent safety, in a manner by which the PTC system shall provide associated warning and enforcement to the extent, and except as, described and justified in the FRA approved PTCDP or PTCSP, as applicable;  Section 5.6.5 Train Movement  Section 5.6.5.3 Wayside Signals  Section 5.6.5.4 Cab Signals  Section 5.6.11.3 Switches Monitored by Signal System  Section 5.6.11.5 Other Monitored Devices  Section 5.6.12 Warning and Enforcement  Section 5.6.12.1 Reactive (Over-speed) Warning and Enforcement  Section 5.6.12.2 Predictive Warning and Enforcement  Section 5.6.12.3 Restrictive Speed Enforcement

6. §236.1005(a)(3) – As applicable, perform the additional functions specified in the subpart;

7. §236.1005(a)(4)(i) - A derail or switch protecting access to the main line required by §236.1007, or otherwise provided for in the applicable PTCSP, is not in its derailing or protecting position, respectively;  Section 5.6.5.3 Wayside Signals  Section 5.6.11.5 Other Monitored Devices  Section 5.6.12 Warning and Enforcement  Section 5.6.12.2 Predictive Warning and Enforcement

8. §236.1005(a)(4)(ii) – Provide an appropriate warning or enforcement when a mandatory directive is issued associated with a highway-rail grade crossing warning system malfunction as required by §234.105, §234.106, or§234.107;  Section 5.6.8 Malfunctioning Highway Grade Crossing Warning Systems  Section 5.6.12 Warning and Enforcement  Section 5.6.12.2 Predictive Warning and Enforcement

9. §236.1005(a)(4)(iii) – Provide an appropriate warning or enforcement when an after- arrival mandatory directive has been issued and the train or trains to be waited on has not yet passed the location of the receiving train;

PTCIP

 Section 5.6.5.2.1 Track Warrant Control

10. §236.1005(a)(4)(iv) – Provide an appropriate warning or enforcement when any movable bridge within the route ahead is not in a position to allow permissive indication for a train movement pursuant to §236.312;  Section 5.6.5.3 Wayside Signals  Section 5.6.11.5 Other Monitored Devices  Section 5.6.12 Warning and Enforcement  Section 5.6.12.2 Predictive Warning and Enforcement

11. §236.1005(a)(4)(v) – Provide an appropriate warning or enforcement when a hazard detector integrated into the PTC system that is required by §236.1005 (c) of this section, or otherwise provided for in the applicable PTCSP, detects an unsafe condition or transmits an alarm;  Section 5.6.5.3 Wayside Signals  Section 5.6.11.5 Other Monitored Devices  Section 5.6.12 Warning and Enforcement  Section 5.6.12.2 Predictive Warning and Enforcement

12. §236.1005(a)(5) – Limit the speed of passenger and freight trains to 59 miles per hour and 49 miles per hour, respectively, in areas without broken rail detection or equivalent safeguards;  Section 5.6.6.1 Permanent Speed Restrictions  Section 5.6.12 Warning and Enforcement  Section 5.6.12.1 Reactive (Over-speed) Warning and Enforcement

PTCIP

4. Compliance [§236.1011(A)(2)] This section describes how CN will comply with §236.1009(d), which requires the railroad to apply for and receive PTC System Certification from the FRA. It is understood that the PTC System Certification must be received before deploying a PTC system(s) in revenue service.

In addition, this section describes any identified or potential risks or other items that could create or suggest increased difficulty in the successful completion and delivery of the PTC system installation on or prior to the required date. It also identifies general contingency plans to deal with risks.

4.1. PTC System Certification CN is pursuing the installation of a PTC system that is fully interoperable with the other Class 1 freight railroads and is actively engaged in the ITC (Interoperable Train Control) initiative. The PTC system that will be installed on CN track will be based on the same equipment technologies and system architecture as the other ITC affiliated railroads.

4.1.1. Utilization of Existing Type Approval and PTCDP The PTC technology chosen by CN is the same system that has been chosen by most of the Class 1 railroads and is based on the Wabtec Interoperable Electronic Train Management System platform. A common PTCDP for the Wabtec I-ETMS has been submitted by a number of Class 1 railways for review and approval by the FRA. The FRA has granted a Type Approval FRA-TA-2011-02 [10] for the I-ETMS platform described in the PTCDP. Accordingly, CN is referencing the I-ETMS Type Approval for the PTCDP that has been granted Type Approval in accordance with §236.1013,in compliance with the requirements of §236.1009(b)(2) in this PTCIP revision. CN will utilize this Type Approval as part of the PTCSP for its PTC system certification request.

CN will identify clearly and explain in its PTCSP any and all variances between the CN proposed PTC system implementation and the I-ETMS platform Type Approval or PTCDP.

4.1.2. Certifying the Validity of Type Approval Section §236.1013(c) in the final rule states, "each Type Approval shall be valid for a period of 5 years, subject to automatic and indefinite extension provided that at least one PTC System Certification using the subject PTC system has been issued within that period and not revoked." It is CN‟s intent to achieve PTC system certification within the 5 year window provided in the rule.

4.1.3. Handling of Unique Aspects of the PTCDP and Type Approval At the time of submission of this PTCIP, CN does not foresee any variances in technology or application from the standard Wabtec I-ETMS based PTC systems used by the other Class 1 freight railroads. Based on the decision to utilize a standard implementation of the I-ETMS based PTC system, CN is not documenting or submitting any unique PTC system aspects as a variance to the Type Approval.

PTCIP

CN has participated in a detailed review of the I-ETMS PTC product with Wabtec technical resources to ensure that the system will provide the functionality required to be successfully deployed on the CN network. In addition to the detailed technical reviews, CN utilizes the TMDS CAD system that was also designed and developed by Wabtec. A single source supplier of both the CAD and PTC office computer systems will help ensure successful integration of the PTC system on the CN network.

Throughout the PTC development and implementation process, CN will keep the FRA fully advised of any issues or circumstances that may develop that would require CN to implement a variance to the standard I-ETMS based PTC platform. This is to ensure that CN maintains compliance with PTC safety certification as rollout of our PTC implementation progresses. If required, RFA(s) will be submitted in accordance with §236.1021 to cover implementation changes.

4.1.4. Deliverables As part of our PTC System Certification process, CN will supply the following deliverables to the FRA: 1. PTC Implementation Plan (PTCIP) 2. PTC Development Plan (PTCDP) Type Approval Number 3. Full description of any variances to the PTCDP or Type Approval 4. PTC Safety Plan (PTCSP)

As required by Section §236.1015 of the final rule, CN will include the following as part of our PTCSP documentation: a. Type Approval reference and copy of approved PTCDP. b. Documentation of installed PTC system variances from system covered by Type Approval of approved PTCDP. c. Human factors analysis of the installed system. d. Hazard log of all safety related hazards. e. Description of safety assurance concepts f. Risk assessment of the as-built system g. Hazard mitigation analysis, h. Description of safety assessment and verification and validation processes i. Description of railway employee training plan j. Procedures and test equipment for employees to operate and maintain system safety through all phases of its life cycle k. Configuration and revision control measures l. Test plans and reports for system configuration and post-implementation testing. m. System operations and maintenance documentation, including warnings and labels; Maintenance and failure records and management.

PTCIP

n. Safety analysis of potential for incursion into established work zone o. Enforcement of integrated hazard detectors. p. Description of how system enforces authorities and signal indications where different from specifics in Part 236 Subpart I. q. Other documents as required by regulation or deemed necessary to support certification of the CN PTC system (e.g., rerouting plan, security requirements).

4.2. Risk Assessment Successful implementation of PTC on CN can be impacted by a number of different risk factors. These risks could create difficulty in completing PTC systems deployment by the 31 December, 2015 completion date set by the FRA or impact the ability of the system to provide all of the required functionality.

To help ensure successful PTC deployment, CN has implemented a risk management process to identify, mitigate, and monitor risks that could create or suggest increased difficulty in the successful completion and delivery of the PTC system installation on or prior to the required date.

This risk management process:  identifies risks to meeting the goals and objectives of CN‟s PTC deployment  predicts the consequences associated with the identified risks;  implements risk mitigation strategies;  monitors risk status; and  establishes contingency plans. This following summary of risks provides a general description of the principal risks that CN believes could impact successful implementation of PTC and is not intended to be an all- inclusive list of every conceivable impediment that could be encountered. CN will maintain the risk management process through which additional risks may be identified and existing risks may be closed as PTC installation progresses.

The sections below provide a summary of identified risks to CN‟s completion and delivery of PTC installation on or prior to December 31, 2015.

4.2.1. Performance Risks Performance Objective 1: Enhance system safety, with particular focus on the prevention of train-to-train collisions, over-speed derailments, incursions into established work zone limits, and movement of trains through improperly-positioned switches.

Table 3 Performance Risk 1 Risk Description Predicted Consequences Risk Mitigation PTC system does not deliver expected system safety benefits:  Does not prevent train to train collisions  PTC system cannot be  Follow system development  Does not prevent overspeed derailments deployed without methodology that captures PTC

PTCIP

Risk Description Predicted Consequences Risk Mitigation  Does not prevent incursions into modification of system system requirements and provides established work zone limits behaviour. traceability of those requirements  Does not prevent movement of trains  PTC system cannot be throughout the system life cycle. through improperly positioned switches deployed without re-  Rigorous safety program at all levels  Creates additional safety hazards that assessment of achieved of system development. reduce system safety safety levels. Methodologies and activities as  An acceptable level of safety is not  Deployed PTC system required by 49CFR§236.1015 will maintained in the development, cannot obtain FRA be followed in the PTCSP. functionality, architecture, installation, Certification implementation, inspection, testing,  Schedule delay operation, maintenance, repair, and  PTC system does not achieve modification of the PTC technologies to expected results for PTC be deployed. preventable incidents

Performance Objective 2: CN will maintain acceptable levels of operation on subdivisions operating under PTC. Table 4 Performance Risk 2 Risk Description Predicted Consequences Risk Mitigation CN incurs unacceptable train delays resulting from PTC operation  PTC implementation and/or system  Railroad incurs unacceptable  Reliability program initiated to design introduces inefficiencies train delays as a result of monitor, report, and improve o wireless communication-related PTC reliability of equipment. delays  PTC deployment is delayed  Identify and reach agreement with o Inefficient train operation resulting until productivity issues are additional potential tenants for from braking algorithm resolved equipping with PTC equipment.  Reduction in efficiency resulting from  Railroad incurs significant  Monitor effectiveness of training – running unequipped trains through PTC revenue penalties caused by quality improvement program in place. equipped territory because service performance issues  System development effort focusing (a) Locomotives operating with PTC  Customers select alternate on high technical risk areas to identify equipment installed but with shipping options for and mitigate potential system design equipment outages products, potentially and implementation-related (b) trains not PTC-equipped. including TIH shipments contributions to decreased productivity  Reduction in efficiency of personnel o Ineffective human factors design for PTC equipment  Ineffective and/or insufficient training of personnel

Contingency Plan: Existing method of operation can be maintained during/after PTC installation until acceptable safety and operational levels have been achieved and FRA Certification has been granted. 4.2.2. Deployment Risks Deployment Objective 1: Enhancements to system safety will be achieved as a PTC vital overlay system is progressively deployed across all portions of the CN network for which PTC deployment is required by 49CFR§236.1005(b).

Table 5 Deployment Risk 1 Risk Description Predicted Consequences Risk Mitigation PTC system progressive installation

PTCIP

Risk Description Predicted Consequences Risk Mitigation is delayed because of  PTC equipment availability  PTC system will not be installed  Develop detailed plans for equipping  Availability of trained installers across all portions of the CN rolling stock, wayside, and office with  Ineffective coordination of network for which PTC required PTC equipment. installation plans result in deployment is required by  Develop detailed training and interference between installation 49CFR§236.1005(b) personnel plans. crews where infrastructure is  Full benefit of safety  Work closely with vendors and other complex and/or working space is enhancements will not be realized railroads in close geographic limited. by required date proximity to minimize risk associated  Installation procedures become  CN may incur Civil Penalties with installation procedures and protracted schedule.  Acts of nature  Establish schedule performance metrics to measure PTC deployment progress. Monitor metrics to identify any potential schedule delays. Take action to avert potential schedule delays.  Deployment targeted to complete highest priority line segments first.

Deployment Objective 2: All required portions of the network to be fully equipped, operational, and interoperable with all tenant railroads by December 31, 2015. Table 6 Deployment Risk 2 Risk Description Predicted Consequences Risk Mitigation All required portions of the network are not fully equipped, operational, and interoperable with all tenant roads by December 31, 2015.  Unable to maintain equipage  PTC system will not be installed  See Risk Mitigation Strategy for schedule across all portions of the CN Coverage risk #1 above.  Delay in initiating PTC operations network for which PTC  Establish clear understanding of  Difficulty and/or delay in deployment is required by technical requirements and schedule establishing required 49CFR§236.1005(b) for interoperability with each tenant interoperability agreements with  Full benefit of safety road. tenant railroads. enhancements will not be realized  Establish performance metrics to  Difficulty and/or delay in by required date measure tenant progress toward achieving required levels of  CN may incur Civil Penalties equipping rolling stock with technical interoperability interoperable PTC equipment.

Contingency Plan: Existing method of operation can be maintained during/after PTC installation until acceptable safety levels have been achieved and FRA Certification has been granted 4.2.3. Compliance Risks Compliance Objective 1: PTC deployment will meet the PTC System Certification performance requirements in C.F.R. §236.1015

Table 7 Compliance Risk 1 Risk Description Predicted Consequences Risk Mitigation The PTC system development does not fully satisfy all of the safety and quality assurance requirements

PTCIP

Risk Description Predicted Consequences Risk Mitigation documented in 49CFR§236.  The methodologies and activities  PTC may not function as required  The methodologies and activities as as required by 49CFR§236.1015 to meet performance requirements. required by 49CFR§236.1015 will be are not applied consistently for  PTC system may not enhance followed for the PTCSP. the PTCSP. safety levels.  CN will ensure that all vendors from  Gaps in the V&V process are  PTC system cannot be deployed whom PTC technologies are to be uncovered that impact the validity without modification of system acquired will have an acceptable of testing results; or, at worst, the behaviour. quality assurance program for both design of the system.  PTC system cannot be deployed design and manufacturing processes. without re-assessment of achieved  Testing and documentation process safety levels. audits are conducted periodically with  Deployed PTC system cannot vendors obtain FRA Certification  Schedule delay

Contingency Plan: Existing method of operation can be maintained during/after PTC installation until acceptable safety levels have been achieved and FRA Certification has been granted

PTCIP

4.2.4. Technical Risks Technical Objective 1: PTC system as deployed successfully provides the required interoperability between CN and its tenants.

Table 8 Technical Risk 1 Risk Description Predicted Consequences Risk Mitigation Interoperability between CN and its tenants is not achieved.  Unsuccessful in deploying  PTC system will not be installed  Establish organizational structure to interoperable radio and messaging across all portions of the CN facilitate communication and technology network for which PTC coordination between host and tenant  Semantic incompatibility between deployment is required by roads railroads 49CFR§236.1005(b)  CN participates in industry  Full benefit of safety organizations to establish PTS system enhancements will not be realized standards to achieve interoperability by required date by working collaboratively on  CN may incur Civil Penalties requirements definition,  Operational penalties incurred on system/component design, and product key service corridors due trains testing to deploy interoperable, operating with failed PTC common technology. equipment.  Testing will ensure that implementations conform to industry standards.  Interoperability testing will be conducted.

Contingency Plan: Existing method of operation can be maintained during/after PTC installation until acceptable safety levels have been achieved and FRA Certification has been granted

PTCIP

5. Interoperability [§236.1011(a)(3)]

This section provides a description of how CN‟s PTC system will provide for interoperability as defined by 49CFR Part 236 Subpart I between CN and the following railroads with which CN has a host or tenant relationship excluding Class II and III railroads, which as defined in 236.1006(b)(4) are permitted to operate on PTC-operated track with non-PTC equipped locomotives (subject to certain FRA conditions):

 Amtrak  Burlington Northern Santa Fe  CSX Transportation Inc.  Canadian Pacific  Kansas City Southern  Metra  Norfolk Southern Railway Company  Union Pacific Railroad

5.1. Railroad Agreement Provisions Relevant to Interoperability [§236.1011(a)(3)(i)] An ITC collaboration agreement was executed by and amongst several railroads wishing to achieve Positive Train Control system interoperability through, in part, the development of an interoperable train control system which would enable locomotives of one participant to transition at track speed to the control of another participant. The collaboration agreement includes a list of interoperability requirements mutually agreed-upon by the parties:  Definition and adoption of uniform interface standards;  Definition, adoption and implementation of AAR standard communications protocols and interoperability standards;  Definition, adoption, and implementation of AAR standard common office- locomotive communications protocols and message formats;  Definition, adoption, and implementation of a common onboard Human Machine Interface, allowing an engineer from any of the participant‟s roads to utilize the system on any participant‟s locomotives on territory for which the engineer is qualified;  Adoption of a coordinated plan for configuration management of the interoperable PTC onboard executable software;  Agreement on use of radio spectrum in the 220MHz band for data radio communications between the locomotive and wayside and the locomotive and back office;  Agreement to acquire, develop and deploy all of the technical capabilities required to permit the use of shared communications infrastructure; and

PTCIP

 Definition and adoption of standards allowing each participant‟s locomotive engineer, at the start of a trip, to initialize the interoperable onboard system with the back offices of participants‟ PTC systems which may be traversed during the trip to support all interoperability scenarios which will be encountered on the line- of-road with respective locomotive fleets and run-through operations. The ITC collaboration agreement chartered the formation of various technical working committees, each dedicated to some technical aspect of PTC interoperability. Participation in the technical working committees was expanded beyond the chartering roads to include any railroad that is planning to implement an interoperable PTC system and wishing to participate.

CN is an active participant on many of the ITC technical teams although not formally party to the ITC collaboration agreement at this time. Through technical team activities, and also through engagements with the principal suppliers of PTC equipment that are party to the ITC development effort, CN is aware of the developments taking place, is confident that there is no impediment to adopting the standards and technology arising from this effort in our organization, and will be able to achieve interoperability.

CN has exchanged Letters of Understanding with each of its passenger tenant carriers (Metra and Amtrak) who are required to install and operate PTC, as well as with all other Class 1 Railways. The Letters of Understanding establish agreement between CN and these parties in the following areas:

 Implementation of PTC technical solutions which meet the requirements of interoperability as defined in §236.1003(b);  Participation in a PTC testing program to verify functionality and interoperability; and  Exchange of technical information needed to implement PTC in accordance with applicable FRA requirements. Copies of the memorandum of understanding letters are attached in Appendix D.

5.2. Types of Interoperability CN will achieve interoperable PTC operations on with its tenant and host railroads which operate PTC systems in one of three technical methods.

5.2.1. Native Interoperability CN and its interoperability partner both install and operate the I-ETMS on their respective locomotives, office, and wayside. I-ETMS provides for full functionality for any equipped locomotive, regardless of ownership, with any office or wayside correspondingly equipped. Interoperability is achieved through native operation of I-ETMS without the need for data, function, or human-machine interface (HMI) translation. Interoperable communications are achieved through adoption of the common communications and message protocols, and application behaviour specifications described in ITC interoperability requirements. I-ETMS encompasses the methods of operation and rules of both CN and its interoperability partner and accommodates any differences in the data provided by back office systems. I-ETMS and its operations are fully described in the Interoperable Electronic Management System

PTCIP

Positive Train Control Development Plan. Railroads with which CN will conduct interoperable PTC operations in this manner are as follows:  Amtrak  Burlington Northern Santa Fe  CSX Transportation Inc.  Canadian Pacific  Kansas City Southern  Metra  Norfolk Southern  Union Pacific Railroad

5.2.2. Onboard Functional Interoperability CN and its interoperability partner install and operate different systems on their respective locomotives, office, and wayside. However, the locomotive onboard system of each is able to interoperate with the office and wayside infrastructure deployed on the other‟s property. Currently, CN does not have any interoperability partners that operate in this fashion.

5.2.3. Unequipped Operation Some of CN interchange partners may operate their unequipped locomotives on CN PTC lines where FRA regulations allow. Although no technical form of interoperability is required or exists, such operations will be conducted as prescribed in §236.1029 and will require procedural coordination amongst CN and its interchange partner. Railroads with which CN will interchange and allow unequipped operation on its PTC lines are as follows:

 IANR – Iowa Northern Railway  MSE – Mississippi Export  CG – Central Gulf Railway  C&J – C&J Railroad Company (Mississippi Delta Railroad)  IHB – Indiana Harbour Belt Railway  HESR – Huron and Eastern Railway  TSBY – Tuscola and Saginaw Bay Railway Company  CSSSB – Chicago South Shore and South Bend Railroad  WSOB – Wisconsin and Southern Railway

All of the Class II/III tenant railways identified above, with the exception of the Wisconsin and Southern Railway, operate trains for less than 20 miles on CN track.

Wisconsin and Southern Railway trains that operate for more than 20 miles on CN‟s Waukesha subdivision will be required to be PTC equipped by 31 December 2020 as specified in 49 CFR §236.1006(b)(4). Initial discussions have been held between CN and the WSOR and CN will continue to work with the WSOR to ensure compliance with the requirements for PTC operations by 2020.

5.3. Technology Applicable to Interoperability [§236.1011(a)(3)(ii)]

PTCIP

CN and its interoperability partners utilize methods in three areas to obtain and maintain interoperability of its PTC system(s):

5.3.1. Technical interoperability Technical interoperability is achieved through the common use of documented interface definitions. These definitions include one or more radio protocols (220MHz) and hardware interfaces to radio equipment, a common standard messaging protocol (ITC Messaging), and standard data element and application message format and content definitions (I-ETMS interface control documents). Use of and compliance with these common interface definitions ensures the ability to exchange data messages between interoperable system components.

5.3.2. Semantic Interoperability Semantic interoperability is achieved through the common use of documented system behavioural specifications. In the current ITC architecture, standard application-level specifications define the behaviour of the interoperable office, locomotive, and wayside segments. Use of and compliance with these common behavioural specifications ensures each interoperable system segment properly interprets and acts upon exchanged data messages.

5.3.3. Organizational interoperability Organizational interoperability is achieved primarily through industry-wide forums, such as committees chartered by ITC and AAR. Technical teams operating under both the ITC and AAR charters are tasked with developing and maintaining the common technical standards in the areas of technical and semantic interoperability described above. These teams have worked to establish a baseline level of interoperability required for industry-wide PTC implementation. The teams will work in perpetuity to provide configuration management and ensure that interoperability is maintained as the interoperable PTC system(s) are enhanced. ITC and AAR teams also work to establish organizational interoperability in the areas of interchange and infrastructure sharing. Finally, CN has designated a liaison to ensure organization communications on PTC interoperability matters with each of its tenant railroads.

5.4. Obstacles to Interoperability [§236.1011(a)(3)(iii)]

As a hosting railroad, CN foresees no obstacles to achieving full interoperability with any and all tenant railroads that operate lead locomotives equipped for PTC certified as conforming to the specifications being established by the ITC consortium, and that also exchange the requisite information for operating a train as established by the ITC consortium.

As a tenant railroad, CN also foresees no obstacles to achieving full interoperability with any and all hosting railroads that operate a wayside equipped for PTC certified as conforming to the specifications being established by the ITC consortium, and that also exchange the requisite information for operating a train as established by the ITC consortium.

CN intends to subject its PTC back office, wayside infrastructure and locomotive equipment for certification or install equipment already type-certified for interoperability as appropriate.

PTCIP

For Class 2 and Class 3 tenant railroads that are not required to install PTC (per 49CFR Part 236), operation of PTC un-equipped trains shall only be permitted in compliance with §236.1006 (b)(4). In these cases, CN intends to mitigate risk by taking one of the following actions:

 Dispatch the train through CN track as a PTC-unequipped train, conforming to all the restrictions prescribed by §236.1029.  Enforce a requirement that the train have a functional PTC-equipped locomotive in the lead while operating on CN PTC-controlled track.  Deny the unequipped train access to CN PTC-controlled track.

All tenant railroads that are required to install PTC (per 49CFR Part 236), will be expected to have a functional PTC-equipped locomotive in the lead while their train is operating on CN PTC-controlled track. In cases where a tenant railroad that is required to install PTC wishes to operate a train on CN PTC-controlled track and the train has a non-functional PTC- equipped locomotive, CN intends to mitigate risk by taking one of the following actions:  Dispatch the train through CN track on an exception basis as a PTC-unequipped train, conforming to all the restrictions prescribed by §236.1029.  Realigning or re-consisting the motive power so that the train has a PTC-equipped locomotive in the lead, including supplying, if necessary, a CN PTC-equipped lead locomotive.  Denying access to PTC-controlled track.

In cases where a tenant railroad that is required to install PTC wishes to operate a train on CN PTC-controlled track and the train does not have a PTC-equipped locomotive, CN intends to mitigate risk by taking one of the following actions:  Until 31 December 2015, dispatch the train through CN track on an exception basis as a PTC-unequipped train, conforming to all the restrictions prescribed by §236.1029.  Realigning or re-consisting the motive power so that the train has a PTC-equipped locomotive in the lead, including supplying, if necessary, a CN PTC-equipped lead locomotive.  Denying access to PTC-controlled track.

PTCIP

6. Designating Track as Main Line or Non-Main Line [§236.1011(a)(8)] This section details which track segments CN considers main line and non-main line track as well as those track segments for which CN is requesting Mainline Track Exclusion Addendums as defined §236.1019.

6.1. CN Network Descriptions The following sections provide descriptive information on some of the unique aspects of the CN Network as it applies to this PTC Implementation Plan.

6.1.1. EJ&E Acquisition On February 1st, 2009, CN completed its acquisition of the principal lines of the EJ&E. The closing follows the Jan. 23, 2009, effective date of the Surface Transportation Board's (STB) Dec. 24, 2008, decision approving the transaction. Since completing the transaction, CN has followed a measured, step-by-step integration of the acquired EJ&E lines to ensure a safe, efficient combination of the two rail operations. The EJ&E runs in an arc around the City of Chicago from Waukegan, Ill., on the north, to Joliet, Ill., on the west, to Gary, Ind., on the southeast, and then to South Chicago.

As part of the PTC planning process, CN has included the acquired EJ&E assets and has applied the same PTC evaluation process to the acquired assets that has been implemented on all other CN tracks. Due to the date of the transaction, CN has limited overall traffic volume data available for the EJ&E Subdivisions for 2008. In addition, subsequent to the acquisition, some CN traffic was re-routed onto the EJ&E lines. For these reasons it was decided that use of 2008 traffic volumes for MGT would not be representative of expected traffic volumes under CN operations. Volumes of TIH/PIH shipments were available for 2008 and these values were slightly higher than the 2009 values so the PTCIP uses 2008 TIH/PIH shipment volumes for the EJ&E lines. CN completed the integration of EJ&E traffic into CN data systems on July 1st, 2009, giving CN a full 6 months of traffic data for the four EJ&E subdivisions. To calculate traffic volume data (MGT) for the CN PTCIP we have used the available 2009 traffic volume data, prorated for a full 12-month period, as the basis for evaluating EJ&E lines against the main line criteria as well as for risk ranking for PTC system implementation for the acquired EJ&E subdivisions (Matteson, Leithton, Lakefront, Illinois River).

6.1.2. P&I Railroad The P&I Railroad is jointly owned segment between MP 0.0 Burlington Jct. and MP 14.0 P&I Jct. CN, BNSF, and CSX are the owners, UP has trackage rights on it. All CN traffic operating on the Bluford Subdivision uses the P&I Railroad between MP 0.8 Metropolis Jct. and MP 4.1 Chiles Jct. CN traffic data does not contain accurate freight tonnage information on the portion of the P&I RR from 4.1 to 14.0. Further investigation is ongoing to get detailed traffic volumes to perform a final assessment of PTC requirements but until this information is available, CN has included the P&I RR in its entirety as part of our PTCIP.

PTCIP

6.1.3. Once final information is available, if there is a need to adjust the PTC implementation plan on the P&I Railroad, this will be requested by a RFA.CN Network Changes – 2008 to 2010 There have been a number of changes to the nomenclature and organizational structure of CN track segments between 2008 and 2010. Therefore when reviewing data for determination of main line vs non-main line track segments it has been necessary to convert some data from 2008 track segment nomenclature to the current 2010 track segment subdivision names. For the purpose of this PTCIP document, all track segment and subdivision nomenclature is based on the current CN network organizational structure. Specific variances from 2008 to 2010 are as follows:

a) Shelby Subdivision – The Shelby subdivision was created in 2009 when the southernmost portion of the Fulton Subdivision (MP 387.9 to 396.8) and the northernmost portion of the Yazoo Subdivision (MP 13.1 to 5.4) were consolidated and re-named the CN Shelby Subdivision. The newly formed Shelby subdivision is the primary route for CN freight traffic and bypasses Memphis from the Fulton Subdivision on the north to the Yazoo Subdivision on the south. All data used for main line track segment determination as well as weighted risk ranking data for line segment prioritization is based on the 2008 data for the respective portions of the Fulton and Yazoo Subdivisions.

b) Marquette Range Subdivision – The Marquette Range Subdivision in Wisconsin was created in 2009 when the Ore and L‟Anse Subdivisions were combined. As unique data is available for both Ore and L‟Anse Subdivisions for 2008, each of these have been evaluated separately for the purpose of main line segment determination but any future PTC evaluation of this track will be performed under the new combined Marquette Range Subdivision.

c) Manistique Subdivision – The Manistique Subdivision in Wisconsin was extended in November of 2009 to include all track that was previously part of the Marinette Subdivision. As unique data is available for both the Manistique and Marinette subdivisions for 2008 each of these have been evaluated separately for the purpose of main line segment determination but any future planning or evaluation of this track will be under the new combined Manistique Subdivision.

d) Grenada Subdivision – A large portion of the Grenada Subdivision was sold by CN in 2009. The remaining northernmost portion of the Subdivision (MP 397.47 to 403.00) has been added to the Memphis Subdivision and will be evaluated as part of the Memphis subdivision for purpose of main line track segment determination. The remaining southern portion of the Grenada Subdivision (MP 703.8 to MP 727.2) has been renamed as the Canton Subdivision and any future planning or evaluation of this track will be under this name.

6.2. CN Main Line Track Segments For the purposes of PTC planning and evaluation, CN has chosen to define a line segment as a subdivision. This decision is based on the following factors:

PTCIP

 CN information and data acquisition systems are aligned with existing subdivision boundaries and therefore provide natural delineation of where PTC evaluation criteria can be readily segmented,

 Train and maintenance operations are aligned with existing subdivision boundaries which will facilitate PTC implementation if deployment schedules and targets are based on a subdivision segmentation of track,

 Subdivision based delineation of track is the common method of segmentation of capital and resources for capital and operating programs. Segmentation of track using different criteria will unnecessarily complicate the specification and tracking of PTC implementation activities,

 Operating corridors between key terminals typically align well with subdivision boundaries which makes them a logical segmentation for PTC project evaluation.

In 2008, CN‟s U.S. network included 82 subdivisions of track (including EJ&E acquisition) All of these subdivisions were reviewed to determine if they qualified as main line track segments under the RSIA and 49CFR§236.1003 PTC regulations. Each of the 82 CN subdivisions were evaluated according to the main line track definitions as included in 49CFR§236.1003(b) and §236.1005(b)(1)(i and ii).

§236.1003 (b) Definition of Main Line: “Main line means, except as provided in §236.1019 or where all trains are limited to restricted speed within a yard or terminal area or on auxiliary or industry tracks, a segment or route of railroad tracks: (1) Of a Class I railroad, as documented in current timetables filed by the Class I railroad with the FRA under §217.7 of this title, over which 5,000,000 or more gross tons of railroad traffic is transported annually; or (2) Used for regularly scheduled intercity or commuter rail passenger service, as defined in 49 U.S.C. 24102, or both. Tourist, scenic, historic, or excursion operations as defined in part 238 of this chapter are not considered intercity or commuter passenger service for purposes of this part.” “§236.1005 Requirements for Positive Train Control systems (b) PTC system installation. (1) Lines required to be equipped. Except as otherwise provided in this subpart, each Class I railroad and each railroad providing or hosting intercity or commuter passenger service shall progressively equip its lines as provided in its approved PTCIP such that, on and after December 31, 2015, a PTC system certified under §236.1015 is installed and operated by the host railroad on each: (i) Main line over which is transported any quantity of material poisonous by inhalation (PIH), including anhydrous ammonia, as defined in §§171.8, 173.115 and 173.132 of this title; (ii) Main line used for regularly provided intercity or commuter passenger service, except as provided in §236.1019…”

PTCIP

Each of the CN subdivisions were reviewed based on the main line track criteria defined in the RSIA and 49CFR§236.1003(b) (1) & (2). Using the 5MGT and regularly scheduled commuter or inter-city passenger train criteria, there are 48 CN subdivisions that qualify as main line track segments. The basic steps used to evaluate each subdivision for qualification as a main line track segment were as follows:

1) Include four subdivisions acquired as part of EJ&E transaction, using prorated 2009 MGT and TIH traffic volume information,

2) Evaluate all subdivisions using 2008 MGT traffic volumes. When there were multiple measurement sections for MGT volumes within the subdivision the weighted average of MGT volumes was used and compared to the 5 MGT threshold for defining a main line track segment. Weighted average was based on the following formula:

Weighted Avg MGT = Sum of (section MGT x section miles) / Subdivision Miles

3) For all subdivisions that fell below the 5MGT threshold using the weighted average formula and where multiple MGT measurement sections were available, each section within the subdivision were evaluated to determine if any sections exceeded 5 MGT. The portions of the subdivision that exceeded 5 MGT were included on the list as main line track segments.

4) All subdivisions were evaluated for the presence of regularly scheduled commuter and inter-city passenger trains. Any subdivision or section of a subdivision with passenger traffic is included as a main line track segment.

5) Subdivisions and subdivision sections that were identified as main line were reviewed to validate and identify portions that fell within yard limits or restricted speed operations. Main line segment mileages were adjusted to reflect these adjustments.

6) Main line track segments that met the criteria for PTC exclusion based on the MTEA exclusion criteria found in 49CFR§236.1019 were identified and reviewed with the appropriate passenger train operators. With the concurrence of the passenger train operators, these segments of track have been submit for MTEA exclusion from PTC installation and were removed from the CN main line track segment list.

6.2.1. CN Subdivisions Exceeding 5 MGT in 2008 The subdivisions identified in the table below had either 2008 weighted average traffic volumes exceeding the 5 MGT threshold or segments of the subdivision that had peak traffic volumes that exceeded the 5 MGT threshold.

Table 9 CN Subdivisions with Traffic Volumes over 5 MGT in 2008 Subdivision 2008 Traffic Data Main Line - Mileages Over 5 MGT

Avg Peak TIH/PIH Passenger From From To To Route MGT MGT Cars Trains/Day MP Station MP Station Miles Baton Rouge 11.5 15.0 22,815 0 364.8 Baton Rouge Jct 444.2 Orleans Jct 79.4 Beaumont 11.2 15.6 8,565 0 0.0 Mobile 185.0 Switchtender 185.0 Bessemer 7.7 19.7 0 0 138.5 Conneaut 0.0 XB 138.5 Bluford 26.1 46.6 11,035 0 40.7 North Siding 0.0 Edgewood Jct 163.6 Cairo 22.4 22.5 4,139 2 363.1 Illinois 405.4 Cairo Jct 42.3

PTCIP

Centralia 29.4 39.5 4,288 6 247.2 Sandoval Jct 363.1 Illinois 115.9 Champaign 39.6 51.5 9,344 6 124.1 Leverett Jct 247.2 Sandoval Jct 123.1 Cherokee 5.7 6.3 29 0 381.2 Tara 508.8 Sioux City 127.6 Chicago 39.7 43.8 8,866 6 14.5 Kensington 124.1 Leverett Jct 109.6 Dubuque 11.6 14.1 649 0 115.1 East Jct 272 Hilltop 156.9 Elsdon 26.0 26.0 6,655 2 19.3 Blue Island 36.1 Griffith 16.8 Flat Rock 7.8 13.2 1,447 0 11.1 FN 39.8 Diann 28.7 Flint 42.4 53.4 8,509 2 178.6 Emmett St. 334.2 Port Huron 155.6 Fox River 9.1 20.0 27 0 208.1 Neenah North 1.4 Green Bay 36.3 Freeport 9.9 11.8 1,997 16 2.9 Cermak 115.1 East Jct 112.2 Fulton 40.1 50.2 10,846 2 40.7 North Siding 387.9 Leewood 120.3 Gilman 2.4 5.5 589 0 81.1 Gilman 110.0 Gibson City 28.9 Hammond 5.4 5.4 4,524 0 43.7 Hammond 0.0 Baton Rouge Jct 43.7 Holly 7.0 9.0 3,399 6 33.3 Waterford 67.0 Durand 33.7 Iron Range 22.0 22.3 0 0 0.0 Two Harbors 74.2 Iron Jct. 74.2 Joliet 4.3 6.8 4,215 16 7.9 Lemoyne XX Summit 1.0 Leithton 19.3 43.5 2,429 0 65.5 Rondout 101.8 E. Bridge Jct 63.7 Manistique 6.5 6.9 0 0 493.3 Soo Yard 385.9 Gladstone 153.8 Marinette 8.1 8.7 0 0 120.7 Gladstone 1.4 Green Bay 119.3 Matteson 11.5 44.4 5,971 0 101.8 E. Bridge Jct 45.4 Kirk Yard Jct 47.1 McComb 24.2 31.2 17,852 2 727.2 Jackson 904.4 Mays Yard 177.2 Memphis 15.2 49.8 11,305 2 391.8 Memphis 396.8 East Jct 5.0 Minneapolis 5.7 5.7 24 0 308.5 Owen 432.1 Withrow 123.6 Minntac 39.3 39.3 0 0 0.0 Wolf 8.0 South Minntac 8.0 Missabe 36.3 74.5 729 0 0.5 Duluth Docks 68.6 Largo 68.1 Mt. Clemens 17.3 17.3 6,753 0 55.6 Tappan 4.6 Milwaukee Jct 51.0 Neenah 52.9 57.1 2,590 0 158.4 Shops Yard 247.0 Hoover 88.6 Ore 13.7 13.8 0 0 113.2 Algoma Jct 174.4 Partridge 61.2 P&I RR 46.6 46.6 2,689 0 0.0 Burlington Jct 14.0 P&I Jct 14.0 Peoria 5.1 8.0 2,247 0 119.2 Mattoon 79.1 Decatur Jct 40.1 Rainy 38.5 44.5 1,451 0 10.7 Nopeming Jct 165.2 Ranier 154.5 Shelby 20.9 24.0 10,846 2 390.0 Aulon 13.1 Lakeview 16.6 Shore Line 16.2 41.9 2,964 6 54.8 Milwaukee Jct 0.0 Manhattan Jct 54.8 South Bend 42.3 51.9 6,016 8 36.1 Griffith 178.6 Emmett St. 142.5 Sprague (US) 44.7 46.0 1,531 0 1.6 Baudette 45.0 Int. Boundary 43.4 St Louis 8.3 12.9 29 0 70.0 Duquoin 32.9 Lenzburg 37.1 Superior 39.8 55.2 2,277 0 247.0 Hoover 480.3 Carson 233.3 Valley 3.2 5.5 863 0 63.3 Junction City 49.9 Wisconsin Rapids 13.4 Waterloo 9.7 9.7 463 0 272.0 Hilltop 381.2 Tara 109.2 Waukesha 52.3 61.6 2,508 21.5 15.5 Tower B12 158.4 Shops Yard 142.9 Yazoo 45.2 48.4 9,099 2 13.1 Lakeview 218.6 Jackson 205.5

6.2.2. Subdivision Segments Exceeding 5 MGT Included in the preceding table are a number of subdivisions where 2008 traffic volumes exceed the 5 MGT threshold for only a portion of the subdivision track miles. The subdivision portions that exceed the 5 MGT threshold are identified as follows:

1. Beaumont Over 5 MGT from MP 3.9 Bayshore Jct to MP 185.0 Switchtender – 181.1 miles 2. Chicago Over 5 MGT from MP 14.5 Kensington to MP 124.1 Leverett Jct. – 109.6 miles 3. Elsdon Over 5 MGT from MP 19.3 Blue Island to MP 36.1 Griffith – 16.8 miles 4. Gilman Over 5 MGT from MP 81.1 Gilman to MP 110.0 Gibson City – 28.9 miles 5. Holly Over 5 MGT from MP 67.0 Durand to MP 33.3 Waterford – 33.7 miles 6. Joliet Over 5 MGT from MP 7.9 Lemoyne to MP 36.7 Jackson Street – 28.8 miles 7. Leithton Over 5 MGT from MP 65.5 Rondout to MP 101.8 East Bridge Jct – 63.7 miles 8. Memphis Over 5 MGT from MP 391.8 Memphis to MP 396.8 East Jct. – 5.0 miles 9. McComb Over 5 MGT from MP 727.2 Jackson to MP 904.4 Mays Yard – 177.2 miles 10. Peoria Over 5 MGT from MP 119.2 Mattoon to MP 79.1 Decatur – 40.1 miles 11. St. Louis Over 5 MGT from MP 32.9 Lenzburg to MP 70.0 Duquoin – 37.1 miles

PTCIP

12. Valley Over 5 MGT from MP 49.9 Wisconsin Rapids to MP 63.3 Junction City – 13.4 miles The CN main line segments from section 6.3.1 are updated to reflect the revised mileages for main line track segments.

6.2.3. Subdivisions with Regularly Scheduled Passenger Trains There are a number of CN Subdivisions and portions of Subdivisions that have regularly scheduled passenger traffic (commuter or inter-city). All tracks with regularly scheduled passenger trains qualify as main line and will be included in the summary list of CN main line track segments. These are as depicted in the table below:

Table 10 Line Segments with Regularly Scheduled Passenger Trains Subdivision 2008 Traffic Data Passenger Traffic Limits Avg Peak TIH/PIH Passenger From From To To Route MGT MGT Cars Trains/Day MP Station MP Station Miles Cairo 22.4 22.5 4,139 2 363.1 Illinois 405.4 Cairo Jct 42.3 Centralia 29.4 39.5 4,288 6 247.2 Sandoval Jct 306.0 Carbondale 58.8 Centralia 29.4 39.5 4,288 2 306.0 Carbondale 363.1 Illinois 57.1 Champaign 39.6 51.5 9,344 6 124.1 Leverett Jct 247.2 Sandoval Jct 123.1 Chicago 39.7 43.8 8,866 6 1.5 16th Street 124.1 Leverett Jct 122.6 Elsdon 26.0 26.0 6,655 2 25.2 UP Thornton Jct 31.0 Munster 5.8 Flint 42.4 53.4 8,509 2 178.6 Emmett St. 334.2 Port Huron 155.6 Freeport 9.9 11.8 1,997 16 2.9 Cermak 4.4 Bridgeport 1.5 Fulton 40.1 50.2 10,846 2 269.2 Cairo Jct 380.4 Woodstock 112.8 Grimsby 0.3 0.3 0 2 0.6 Int'l Border 0.0 Bridge 0.6 Holly 7.0 9.0 3,399 6 4.1 Milwaukee Jct 25.8 Pontiac 21.7 Joliet 4.3 6.8 4,215 16 3.5 Bridgeport 36.7 Jackson Street 33.2 McComb 24.2 31.2 17,852 2 727.2 Jackson 908.6 Southport Jct. 181.4 Memphis 15.2 49.8 11,305 2 380.4 Woodstock 394.3 West Jct 13.9 Rouse's Point 0.6 0.6 0 2 1.2 Int'l Border 0.0 Rouse's Point 1.2 Shelby 20.9 24.0 10,846 2 5.4 West Jct 13.1 Lakeview 7.7 Shore Line 16.2 41.9 2,964 6 54.8 Milwaukee Jct 51.2 CP Vinewood 3.6 South Bend 42.3 51.9 6,016 8 175.5 Gord 176.7 Baron 1.2 South Bend 42.3 51.9 6,016 2 176.7 Baron 178.6 Emmett St. 1.9 St Charles Airline N/A N/A 0 6 1.7 Michigan Ave 2.3 South Branch 0.6 Waukesha 52.3 61.6 2,508 22 15.5 Tower B12 55.7 Antioch 40.2 Y&MV Main N/A N/A 0 2 0.0 Y&MV Jct. 5.4 West Jct 5.4 Yazoo 45.2 48.4 9,099 2 13.1 Lakeview 218.6 Jackson 205.5

6.2.4. Restricted Speed Track Revisions to Line Segment Mileages The mileage limits for the main line segments identified in section 6.2.1 include the full timetable mileage ranges for all tracks included as part of the Subdivision in the CN Operating Timetables. All track within the identified mileage limits has been reviewed to identify any locations where all train operations are limited to restricted speed and would therefore be excluded from being considered main line track as per the definition in 49CFR§236.1003(b).

The following list summarizes all subdivisions that have sections of restricted speed track within the identified main line track sections (excluding any restricted speed track segments with regularly scheduled passenger train operations):

Table 11 Line Segments with Restricted Speed Track Subdivision Main Line - Over 5MGT or Passenger Traffic Restricted Speed Track in Main Restricted

PTCIP

From From To To Route Line Limits Speed Track MP Station MP Station Miles (miles)

Baton Rouge 364.8 Baton Rouge Jct 444.2 New Orleans Jct 79.4 364.8-366.7 1.9 Beaumont 0.0 Mobile 185.0 Switchtender 185.0 0.0-6.7, 90.9-97.0, 181.6-185.0 16.2 Bessemer 138.5 Conneaut 0.0 XB 138.5 138.5-137.6 0.9 Cairo 363.1 Illinois 405.4 Cairo Jct 42.3 404.8-405.4 0.6 Cherokee 381.2 Tara 508.8 Sioux City 127.6 381.2-381.5 0.3 Flat Rock 11.1 FN 39.8 Diann 28.7 14.4-19.2 4.8 Fox River 208.1 Neenah North 1.4 Green Bay 36.3 1.4-243 1.4 Fulton 269.2 Cairo Jct 387.9 Leewood 120.3 40.7-269.2 1.6 Hammond 43.7 Hammond 0.0 Baton Rouge Jct 43.7 0.0-1.0 1.0 Iron Range 0.0 Two Harbors 74.2 Iron Jct. 74.2 0.0-0.4 0.4 Manistique 493.3 Soo Yard 342.7 Gladstone 150.6 342.7-343.3, 491.3-493.3 2.6 Marinette 120.7 Gladstone 1.4 Green Bay 119.3 1.4-4.0, 119.5-120.7 3.8 Matteson 101.8 E. Bridge Jct 45.4 Kirk Yard Jct 47.1 2.0-101.7 3.7 Memphis 380.4 Memphis 396.8 East Jct 16.4 394.3-396.8 2.5 Missabe 0.5 Duluth Docks 68.6 Largo 68.1 0.5-1.0, 6.6-10.7 4.6 Mt.Clemens 55.6 Tappan 4.6 Milwaukee Jct 51.0 4.6-10.0 5.4 Neenah 158.4 Shops Yard 247.0 Hoover 88.6 158.4-160.4 2.0 Ore 113.2 Algoma Jct 174.4 Partridge 61.2 113.2-117.0, 172.2-174.4 6.0 P&I RR 0.0 Burlington Jct 14.0 P&I Jct 14.0 0.0-1.0 1.0 Rainy 10.7 Nopeming Jct 165.2 Ranier 154.5 162.9-165.2 2.3 Shelby 390.0 Aulon 13.1 Lakeview 16.6 395.5-5.4 3.7 Shore Line 54.8 Milwaukee Jct 0.0 Manhattan Jct 54.8 0.0-3.8 3.8 Valley 63.3 Junction City 49.9 Wisconsin Rapids 13.4 49.9-51.0 1.1 Waterloo 272.0 Hilltop 381.2 Tara 109.2 272.1-277.5 5.4 Waukesha 15.5 Tower B12 158.4 Shops Yard 142.9 157.2-158.4 1.2 Yazoo 13.1 Lakeview 218.6 Jackson 205.5 217.2-218.6 1.4

6.2.5. Final CN Main Line Track Segment Mileages The final CN main line track segment mileages have been identified based on the RSIA and 49CFR236 criteria of 5 MGT annual traffic volumes with TIH/PIH traffic or regularly scheduled passenger operations and adjusted to compensate for track that falls within yards or restricted speed operations. Any portions of restricted speed track where there are regularly scheduled passenger operations, have been noted and are either retained as main line track as required by 49CFR§236.1003 or identified as main line track exceptions as permitted under §236.1019 and are summarized for MTEA submission in section 6.4. The table bellows provides a consolidated view of all CN main line track based on the requirements of the RSIA as well as 49CFR§236.1003 and §236.1005(b)(1)(i and ii).

Table 12 CN Main Line Track Segments Subdivision 2008 Traffic Data Main Line - Over 5MGT or Passenger Traffic Restricted Main Speed Line Avg Peak TIH/PIH Passenger From From To To Route Track (Route MGT MGT Cars Trains/Day MP Station MP Station Miles (miles) Miles)

Baton Rouge 11.5 15.0 22,815 0 364.8 Baton Rouge Jct 444.2 Orleans Jct 79.4 1.9 77.5 Beaumont 11.2 15.6 8,565 0 0.0 Mobile 185.0 Switchtender 185.0 16.2 168.8 Bessemer 7.7 19.7 0 0 138.5 Conneaut 0.0 XB 138.5 0.9 137.6 Bluford 26.1 46.6 11,035 0 40.7 North Siding 0.0 Edgewood Jct 163.6 0.0 163.6 Cairo 22.4 22.5 4,139 2 363.1 Illinois 405.4 Cairo Jct 42.3 0.6 41.7 Centralia 29.4 39.5 4,288 6 247.2 Sandoval Jct 363.1 Illinois 115.9 0.0 115.9 Champaign 39.6 51.5 9,344 6 124.1 Leverett Jct 247.2 Sandoval Jct 123.1 0.0 123.1 Cherokee 5.7 6.3 29 0 381.2 Tara 508.8 Sioux City 127.6 0.3 127.3

PTCIP

Subdivision 2008 Traffic Data Main Line - Over 5MGT or Passenger Traffic Restricted Main Speed Line Avg Peak TIH/PIH Passenger From From To To Route Track (Route MGT MGT Cars Trains/Day MP Station MP Station Miles (miles) Miles)

Chicago 39.7 43.8 8,866 6 1.5 16th Street 124.1 Leverett Jct 122.6 0.0 122.6 Dubuque 11.6 14.1 649 0 115.1 East Jct 272.0 Hilltop 156.9 0.0 156.9 Elsdon 26.0 26.0 6,655 2 19.3 Blue Island 36.1 Griffith 16.8 0.0 16.8 Flat Rock 7.8 13.2 1,447 0 11.1 FN 39.8 Diann 28.7 4.8 23.9 Flint 42.4 53.4 8,509 2 178.6 Emmett St. 334.2 Port Huron 155.6 0.0 155.6 Fox River 9.1 20.0 27 0 208.1 Neenah North 1.4 Green Bay 36.3 1.4 34.9 Freeport 9.9 11.8 1,997 16 2.9 Cermak 115.1 East Jct 112.2 0.0 112.2 Fulton 40.1 50.2 10,846 2 40.7 North Siding 387.9 Leewood 120.3 1.6 118.7 Gilman 2.4 5.5 589 0 81.1 Gilman 110.0 Gibson City 28.9 0.0 28.9 Grimsby 0.3 0.3 0 2 0.6 Int'l Border 0.0 Bridge 0.6 0.6 0.6 Hammond 5.4 5.4 4,524 0 43.7 Hammond 0.0 Baton Rouge Jct 43.7 1.0 42.7 Holly 7.0 9.0 3,399 6 4.1 Milwaukee Jct 67.0 Durand 62.9 0.0 62.9 Iron Range 22.0 22.3 0 0 0.0 Two Harbors 74.2 Iron Jct. 74.2 0.4 73.8 Joliet 4.3 6.8 4,215 16 3.5 Lemoyne 36.7 Jackson Street 33.2 0.0 33.2 Leithton 19.3 43.5 2,429 0 65.5 Rondout 101.8 E. Bridge Jct 65.0 0.0 63.7 Manistique 6.5 6.9 0 0 493.3 Soo Yard 385.9 Gladstone 153.8 2.6 151.2 Marinette 8.1 8.7 0 0 120.7 Gladstone 1.4 Green Bay 119.3 3.8 115.5 Matteson 11.5 44.4 5,971 0 101.8 E. Bridge Jct 45.4 Kirk Yard Jct 47.1 3.7 43.4 McComb 24.2 31.2 17,852 2 727.2 Jackson 908.6 Southport Jct. 181.4 0.0 181.4 Memphis 15.2 49.8 11,305 2 380.4 Memphis 396.8 East Jct 16.4 2.5 13.9 Minneapolis 5.7 5.7 24 0 308.5 Owen 432.1 Withrow 123.6 0.0 123.6 Minntac 39.3 39.3 0 0 0.0 Wolf 8.0 South Minntac 8.0 0.0 8.0 Missabe 36.3 74.5 729 0 0.5 Duluth Docks 68.6 Largo 68.1 4.6 63.5 Mt. Clemens 17.3 17.3 6,753 0 55.6 Tappan 4.6 Milwaukee Jct 51.0 5.4 45.6 Neenah 52.9 57.1 2,590 0 158.4 Shops Yard 247.0 Hoover 88.6 2.0 86.6 Ore 13.7 13.8 0 0 113.2 Algoma Jct 174.4 Partridge 61.2 6.0 55.2 P&I RR 46.6 46.6 2,689 0 0.0 Burlington Jct 14.0 P&I Jct 14.0 1.0 13.0 Peoria 5.1 8.0 2,247 0 119.2 Mattoon 79.1 Decatur Jct 40.1 0.0 40.1 Rainy 38.5 44.5 1,451 0 10.7 Nopeming Jct 165.2 Ranier 154.5 2.3 152.2 Rouse's Point 0.6 0.6 0 2 1.2 Int'l Border 0.0 Rouse's Point 1.2 1.2 1.2 Shelby 20.9 24.0 10,846 2 390.0 Aulon 13.1 Lakeview 16.6 3.7 12.9 Shore Line 16.2 41.9 2,964 6 54.8 Milwaukee Jct 0.0 Manhattan Jct 54.8 3.8 51.0 South Bend 42.3 51.9 6,016 8 36.1 Griffith 178.6 Emmett St. 142.5 0.0 142.5 Sprague (US) 44.7 46.0 1,531 0 1.6 Baudette 45.0 Int. Boundary 43.4 0.0 43.4 St Charles Airline N/A N/A 0 6 1.7 Michigan Ave 2.3 South Branch 0.6 0.6 0.6 St Louis 8.3 12.9 29 0 70.0 Duquoin 32.9 Lenzburg 37.1 0.0 37.1 Superior 39.8 55.2 2,277 0 247.0 Hoover 480.3 Carson 233.3 0.0 233.3 Valley 3.2 5.5 863 0 63.3 Junction City 49.9 Wisconsin Rapids 13.4 1.1 12.3 Waterloo 9.7 9.7 463 0 272.0 Hilltop 381.2 Tara 109.2 5.4 103.8 Waukesha 52.3 61.6 2,508 22 15.5 Tower B12 158.4 Shops Yard 142.9 1.2 141.7 Y&MV Main N/A N/A 0 2 0.0 Y&MV Jct. 5.4 West Jct 5.4 0.0 5.4 Yazoo 45.2 48.4 9,099 2 13.1 Lakeview 218.6 Jackson 205.5 1.4 204.1

Non-Main Line Track: CN considers all other auxiliary track, branch lines, industrial sidings, low tonnage spurs and other track not included in the map and table above to be non- main line track.

6.3. Summary of Technical Notes on CN Data As discussed in the preceding text, it was necessary to make a number of decisions and adjustments concerning the data used to determine which track segments that meet the main line track criteria under RSIA and 49CFR§236.1003(b) (1) & (2). Following is a summary of these decisions and adjustments:

PTCIP

1. EJ&E Subdivisions use 2009 traffic volumes (MGT) for July to December prorated for a full 12 month period to determine if they meet the 5MGT threshold for main line track. These subdivisions are the Matteson, Leithton, Lakefront and Illinois River. Data was available for 2008 TIH shipment volumes and this data has been used for TIH calculations.

2. When there were multiple measurement sections for MGT volumes within the subdivision, the weighted average of MGT volumes was used and compared to the 5 MGT threshold for defining a main line track segment. Weighted average was based on the following formula:

Weighted Avg MGT = Sum of (section MGT x section miles) / Subdivision Miles

3. For all subdivisions with weighted averages that fell below the 5MGT threshold and where multiple MGT measurement sections were available, each section within the subdivision was evaluated to determine if any sections exceeded 5 MGT. The portions of the subdivision that exceeded 5 MGT were included on the list as main line track segments.

4. Subdivisions that did not meet the main line criteria (5 MGT) but had passenger train operations on all or a portion of the subdivision had all track mileages with regularly scheduled passenger trains identified as main line.

5. Passenger train volume for all track on a subdivision is based on the average number of daily passenger train movements on the busiest passenger traffic segment of the subdivision.

6. TIH/PIH traffic volumes include both loaded and residue shipments.

6.4. Foreign Owned Line Segments There are a number of segments of track within the identified main line track segments that are not owned and/or dispatched by CN. These line segments are as follows:

1. Shore Line Subdivision – 3.2 miles from MP 50.2 (West Detroit) to MP 47.0 (River Rouge) – Track is owned and dispatched by NS,

2. Dubuque Subdivision – 13.2 miles from MP 168.8 (Portage) to MP182.0 (East Dubuque) – Main track #1 is owned and dispatched by BNSF, main tack #2 is owned by CN but maintained and dispatched by BNSF,

3. Cherokee Subdivision – 22.5 miles from MP 484.9 (LeMars) to MP507.4 (28th Street) – Track is maintained and dispatched by UP but owned by CN.

4. Shelby Subdivision – 2.1 miles from MP 387.9 (Leewood) to MP 390.0 (Aulon) – Track is owned and dispatched by CSX,

PTCIP

These track segments will be carried forward in this PTCIP document as CN track segments for the purpose of PTC risk factor analysis and PTC deployment planning. Actual responsibility for PTC installation will reside with the “host railroad” which as specified in §236.1005(b) is the railroad that has effective operating control over the segment. CN will engage in discussions with the responsible corporate entities to ensure they are aware of the associated PTC requirements as appropriate.

6.5. MTEA Requests This section includes details of the specific track segments that meet the criteria defined in 49CFR236 to qualify as a main line track segment but for which CN is requesting Mainline Track Exclusion Addendums as defined §236.1019. Each MTEA request is detailed separately in the following sections but all have been reviewed in detail with the applicable passenger train operators and all are submit with their full concurrence and agreement. Each MTEA submission provides a summary track description and layout as well as a narrative description of the normal train operations and a reference to the applicable section of 49CFR §236.1019 that the MTEA is requested under.

MTEA requests being applied for by CN are covered by one of the following exception conditions:

1. 49CFR§236.1019 (c)(1)(i) – the track is used for limited operations by at least one passenger railroad with all trains limited to restricted speed,

2. 49CFR§236.1019(c)(3) – not more than four passenger trains per day are operated on a segment of track of a Class 1 freight railroad on which less than 15 million gross tons of freight traffic is transported annually.

The following list provides an overview of the MTEA‟s being requested by CN:

1. Freeport Sub / Chicago Sub / St. Charles Airline to reach Chicago Union Station

a. Freeport Sub from MP 2.9 to MP 2.1,

b. Chicago Sub from MP 2.2 to 1.4,

c. St. Charles Airline from 16th Street interlocking to connection with BNSF Railway at the end of CN‟s ownership approximately 70 feet west of the bascule bridge over the South Branch of the Chicago River (approximately 0.6 miles).

2. Memphis Sub MP 380.4 (Woodstock) to MP 394.3 (Y&MV Junction),

3. Y&MV Main from Y&MV Junction (Memphis Sub) to MP 5.4 (Shelby Sub),

4. Whirlpool Bridge – MP 0.0 to MP 0.6 Grimsby Subdivision

5. McComb Sub from MP 904.4 (Mays Yard) to MP 908.6 (Southport Junction).

6. Rouse‟s Point Sub from MP 1.18 (International Border) to MP 0.0 (Rouse‟s Point)

PTCIP

Each of the track segments identified above have been excluded from the CN main line track segment list and have also been excluded from PTC risk factor evaluation and PTC deployment scheduling (sections 7 & 8 of this document).

Details on all MTEA requests are included in Section 13 of this document.

PTCIP

7. Installation Risk Analysis [§236.1011(a)(4)] This section describes how CN will comply with 49CFR236, Subpart I, §236.1011(a)(4), which requires the deployment of PTC in areas of greater risk to the public and railroad employees before areas of lower risk.

7.1. The Rail Network The CN rail network is illustrated in the map in section 1.

Of the 48 subdivisions that meet the main line track criteria (see section 6) six subdivisions did not have any passenger trains or TIH/PIH traffic (loads or residue) in 2008 and therefore were eliminated from the CN PTC. In addition, the Y&MV Main is covered by an MTEA request, and has therefore been removed from the implementation planning and risk prioritization process. The table below identifies the resulting 41 CN subdivisions that are considered main line track segments requiring PTC and will be evaluated for installation risk prioritization per the model in Appendix B. This table excludes portions of track identified to be within yard limits where all train operations are limited to restricted speed (see section 6).

Table 13 CN Main line Track segments – Excluding MTEA Tracks Subdivision 2008 Traffic Data Main Line - Over 5MGT or Passenger Traffic Restricted Main Line Avg Peak TIH/PIH Passenger From From To To Route Speed Track (Route MGT MGT Cars Trains/Day MP Station MP Station Miles (miles) Miles) Baton Rouge 11.5 15.0 22,815 0 364.8 Baton Rouge Jct 444.2 Orleans Jct 79.4 1.9 77.5 Beaumont 11.2 15.6 8,565 0 0.0 Mobile 185.0 Switchtender 185.0 16.2 168.8 Bluford 26.1 46.6 11,035 0 40.7 North Siding 0.0 Edgewood Jct 163.6 0.0 163.6 Cairo 22.4 22.5 4,139 2 363.1 Illinois 405.4 Cairo Jct 42.3 0.6 41.7 Centralia 29.4 39.5 4,288 6 247.2 Sandoval Jct 363.1 Illinois 115.9 0.0 115.9 Champaign 39.6 51.5 9,344 6 124.1 Leverett Jct 247.2 Sandoval Jct 123.1 0.0 123.1 Cherokee 5.7 6.3 29 0 381.2 Tara 508.8 Sioux City 127.6 0.3 127.3 Chicago 39.7 43.8 8,866 6 2.2 124.1 Leverett Jct 121.9 0.0 121.9 Dubuque 11.6 14.1 649 0 115.1 East Jct 272.0 Hilltop 133.7 0.0 133.7 Elsdon 26.0 26.0 6,655 2 19.3 Blue Island 36.1 Griffith 16.8 0.0 16.8 Flat Rock 7.8 13.2 1,447 0 11.1 FN 39.8 Diann 28.7 4.8 23.9 Flint 42.4 53.4 8,509 2 178.6 Emmett St. 334.2 Port Huron 155.6 0.0 155.6 Fox River 9.1 20.0 27 0 208.1 Neenah North 1.4 Green Bay 36.3 1.4 34.9 Freeport 9.9 11.8 1,997 16 2.9 Cermak 115.1 East Jct 112.2 0.0 112.2 Fulton 40.1 50.2 10,846 2 40.7 North Siding 387.9 Leewood 120.3 1.6 118.7 Gilman 2.4 5.5 589 0 81.1 Gilman 110.0 Gibson City 28.9 0.0 28.9 Hammond 5.4 5.4 4,524 0 43.7 Hammond 0.0 Baton Rouge Jct 43.7 1.0 42.7 Holly 7.0 9.0 3,399 6 4.1 Milwaukee Jct 67.0 Durand 62.9 0.0 62.9 Joliet 4.3 6.8 4,215 16 3.5 Lemoyne 36.7 Jackson Street 33.2 0.0 33.2 Leithton 19.3 43.5 2,429 0 65.5 Rondout 101.8 E. Bridge Jct 65.0 0.0 63.7 Matteson 11.5 44.4 5,971 0 101.8 E. Bridge Jct 45.4 Kirk Yard Jct 47.1 3.7 43.4 McComb 24.2 31.2 17,852 2 727.2 Jackson 904.4 Mays Yard 177.2 0.0 177.2 Memphis 15.2 49.8 11,305 2 394.3 Memphis 396.8 East Jct 2.5 0.0 2.5 Minneapolis 5.7 5.7 24 0 308.5 Owen 432.1 Withrow 123.6 0.0 123.6 Missabe 36.3 74.5 729 0 0.5 Duluth Docks 68.6 Largo 68.1 4.6 63.5 Mt. Clemens 17.3 17.3 6,753 0 55.6 Tappan 4.6 Milwaukee Jct 51.0 5.4 45.6 Neenah 52.9 57.1 2,590 0 158.4 Shops Yard 247.0 Hoover 88.6 2.0 86.6 P&I RR 46.6 46.6 2,689 0 0.0 Burlington Jct 14.0 P&I Jct 14.0 1.0 13.0 Peoria 5.1 8.0 2,247 0 119.2 Mattoon 79.1 Decatur Jct 40.1 0.0 40.1 Rainy 38.5 44.5 1,451 0 10.7 Nopeming Jct 165.2 Ranier 154.5 2.3 152.2 Shelby 20.9 24.0 10,846 2 390.0 Aulon 13.1 Lakeview 16.6 3.7 12.9 Shore Line 16.2 41.9 2,964 6 54.8 Milwaukee Jct 0.0 Manhattan Jct 54.8 3.8 51.0

PTCIP

Subdivision 2008 Traffic Data Main Line - Over 5MGT or Passenger Traffic Restricted Main Line Avg Peak TIH/PIH Passenger From From To To Route Speed Track (Route MGT MGT Cars Trains/Day MP Station MP Station Miles (miles) Miles) South Bend 42.3 51.9 6,016 8 36.1 Griffith 178.6 Emmett St. 142.5 0.0 142.5 Sprague (US) 44.7 46.0 1,531 0 1.6 Baudette 45.0 Int. Boundary 43.4 0.0 43.4 St Louis 8.3 12.9 29 0 70.0 Duquoin 32.9 Lenzburg 37.1 0.0 37.1 Superior 39.8 55.2 2,277 0 247.0 Hoover 480.3 Carson 233.3 0.0 233.3 Valley 3.2 5.5 863 0 63.3 Junction City 49.9 Wisconsin Rapids 13.4 1.1 12.3 Waterloo 9.7 9.7 463 0 272.0 Hilltop 381.2 Tara 109.2 5.4 103.8 Waukesha 52.3 61.6 2,508 21.5 15.5 Tower B12 158.4 Shops Yard 142.9 1.2 141.7 Yazoo 45.2 48.4 9,099 2 13.1 Lakeview 218.6 Jackson 205.5 1.4 204.1

7.2. Risk Factors, Risk Factor Levels, and Risk Factor Weights The risk prioritization model used by CN is based on a risk evaluation methodology that was developed through a cooperative effort between a number of the Class 1 railways working with the Rail Safety group at Battelle. The prioritization model incorporates a basic weighted score approach in which a number of risk factors were assigned integer scores, corresponding with level of risk, ranging from 0 (lowest risk) up to 5 (highest risk) for each of the CN subdivisions to be equipped with PTC. Each risk factor was also assigned a weight, which provided an indication of the “relative importance” of the factor in determining the overall risk ranking. Equation 1 below shows how, for n risk factors, a relative risk score was generated for each subdivision by multiplying the integer score assigned to the subdivision for a given factor (FRi) by the weight assigned to that factor (FWi), and summing the products of the n risk factors.

Equation 1 Relative Risk Score for Subdivision =

In order to perform the above calculation, the following activities were undertaken:

1. Identify risk factors to be included in the risk prioritization model;

2. Estimate the risk factor weights (FWi); includes subjective assessment of risk probability and risk consequence; 3. Define the ranges of data for each of the 6 risk factor levels (0 – 5) that would be used to assign scores to the subdivisions for each risk factor; The lower and upper limits of the data defined for each risk factor level reflect a normalized range as determined by CN;

4. Assign integer scores (FRi) to each subdivision using the criteria defined in #3 above. The FRA Risk Prioritization Methodology for PTC System Implementation includes a list of risk factors, which it identifies as “minimum critical risk factors that must be addressed” in the risk prioritization model. These eight risk factors, which are listed below, correspond with the risk factors identified in §236.1011(a)(5) as minimum factors that will be used in the consideration of the order the track segments will be equipped; CN evaluated these eight risk factors in the risk prioritization model.

PTCIP

7.2.1. Risk Factor 1: Annual Million Gross Ton (MGT)

This risk factor provides an indication of the annual volume of traffic present on a subdivision.

An increased volume of traffic corresponds with increased probability of various types of accidents. This is particularly the case on single track segments, where meets and passes increase as total train traffic increases, creating an increased probability of train-to-train collision. On segments of multiple tracks, meets and passes are not increased to the same extent due to increased capacity. Also, while increased volume would primarily affect the probability component of risk, it is expected that consequences could be slightly increased on subdivisions with higher traffic volume due to the fact that the occurrence of an accident could require re-routing of an increased number of trains and cause greater disruption to commerce than an accident on a less-traveled subdivision, thereby resulting in increased economic costs.

The implementation of PTC is expected to achieve greater reductions in risk for subdivisions with higher MGT levels.

For each subdivision with multiple MGT measurement sections, a weighted average of the MGT levels for each measurement section was used to generate an aggregate measurement for the subdivision (i.e., Weighted Avg MGT = Sum of (section MGT x section miles) / Subdivision Miles). Limits for six factor levels, provided in the table below, were assigned taking the range of subdivision MGT data into account and based on the CN subdivision maximum of 53 MGT per year.

Table 14 Annual MGT Risk Factor Levels Factor Levels for Annual MGT Level Factor Level Lower Limit Upper Limit Level 0 0 MGT <5 MGT Level 1 5 MGT <15 MGT Level 2 15 MGT <25 MGT Level 3 25 MGT <35 MGT Level 4 35 MGT <45 MGT Level 5 45 MGT None

7.2.2. Risk Factor 2: Presence and Volume of Passenger Traffic

This risk factor addresses the rounded daily volume of passenger trains that travel on a subdivision.

Due to the fact that significant value is assigned to injury and loss of life when assessing the consequences of an accident, accidents involving passenger trains generally have greater consequences than those involving freight trains only. Casualties arising from freight train accidents are normally limited to on-board railroad operations personnel whereas accidents involving passenger trains have a greater injury and death toll with the traveling public.

PTCIP

Since the potential for accidents involving passenger trains would clearly increase along with passenger traffic, the installation of PTC is expected to achieve greater reductions in risk for subdivisions with higher volumes of passenger traffic.

For each subdivision, the highest number of passenger trains per day running on one or more of the subdivision segments determined the passenger train level for the entire subdivision. Limits for six factor levels, provided in the table below, were assigned taking the range of passenger train volume data into account and based on the CN subdivision maximum of 22 passenger trains per day.

Table 15 Daily Passenger Train Risk Factor Levels Factor Levels for Presence and Volume of Passenger Traffic Factor Level Lower Limit Upper Limit Level 0 0 Trains/Day 0 Trains/Day Level 1 1 Trains/Day 2 Trains/Day Level 2 3 Trains/Day 5 Trains/Day Level 3 6 Trains/Day 10 Trains/Day Level 4 11 Trains/Day 20 Trains/Day Level 5 21 Trains/Day None

7.2.3. Risk Factor 3: Presence and Volume of TIH/PIH Material (Loads and Residue) Transported

This risk factor provides an indication of the annual volume of TIH/PIH materials that are transported on a subdivision.

The presence of TIH/PIH materials could significantly increase the consequences of an accident, as accidents involving trains carrying TIH/PIH materials may not only cause injury and/or loss of life to those onboard the train and in the railroad right-of-way, but also injury and/or loss of life to the general population in areas beyond the railroad, potentially even miles away from the location of the accident. Depending on the physical nature of the accident (extent of damage to cars, specific materials, chemical phase, flow rates, meteorology, population density in the affected area, etc.), a TIH/PIH release could potentially cause physical harm to large numbers of exposed individuals.

This risk factor primarily affects the consequence component of risk for the reasons described above, and due to the fact that TIH/PIH loads are accounted for in the „Annual MGT Level‟ risk factor (i.e., this risk factor does not represent the presence of additional

PTCIP

traffic), it is estimated that the „Presence and Volume of TIH/PIH Material‟ risk factor does not significantly affect the probability component of risk in this assessment.

The implementation of PTC is expected to achieve greater reductions in risk for subdivisions upon which higher volumes of TIH/PIH loads and residues are transported.

For subdivisions where TIH/PIH materials were transported in 2008, the numbers of cars per year which transport TIH/PIH materials were identified. For those subsections with multiple measurement sections, the maximum number of TIH cars for one or more of the sections was used for the entire subsection. Limits for six factor levels, provided in the table below, were assigned taking the range of CN subdivision TIH/PIH data into account and based on the CN subdivision maximum of 22,815 TIH/PIH cars per year.

Table 16 Annual Car Volume of TIH/PIH Risk Factor Levels Factor Levels for Presence and Volume of TIH/PIH Material (Loads and Residue) Transported Factor Level Lower Limit Upper Limit Level 0 0 Cars/Year 0 Cars/Year Level 1 1 Cars/Year <100 Cars/Year Level 2 100 Cars/Year <1,000 Cars/Year Level 3 1,000 Cars/Year <5,000 Cars/Year Level 4 5,000 Cars/Year <10,000 Cars/Year Level 5 10,000 Cars/Year None

7.2.4. Risk Factor 4: Number of Tracks

This risk factor addresses the presence of multiple main line tracks along portions of a subdivision.

The increased capacity provided by multiple tracks allows for the number of train meets and passes to be reduced. A reduction in the number of meets and passes would be expected to decrease the number of train-to-train collisions, such that the presence of multiple tracks provides a reduction in risk relative to instances of single track. Although traffic volume may tend to be increased on segments with additional capacity provided by multiple tracks, it is estimated that an increase in traffic volume would not be expected to offset the benefits provided by multiple tracks. As the „Number of Tracks‟ risk factor is therefore assumed to affect the frequency with which collisions occur, it primarily affects the probability component of risk.

It is expected that PTC will prevent many of the train-to-train collisions that might have otherwise taken place on segments of single track through warning and enforcement of the limit of authority, while not preventing as many collisions on double track simply due to the fact that not as many collisions would have been expected to occur on double track. As a result, installing PTC on single track is expected to have increased benefit in this respect.

PTCIP

The percentage of double or multi-track (greater than 2 tracks) on a subdivision was calculated using the distance of multiple track and the total distance of the subdivision to be equipped. Since overall risk should decrease with the number of tracks, a subdivision needed to have more than 50% of its route miles as double or multi-track to be classified at the lower risk rating. That is, a subdivision needed to have at least 51% of its route miles as double track to be classified as double track and more than 51% of its route miles as multi- track to be classified as multi-track track. Limits for three factor levels, provided in the table below, were defined at the one track, two track, and greater than 2 track levels and based on the CN subdivision maximum of 4 main tracks.

Table 17 Number of Tracks Risk Factor Levels Factor Levels for Number of Tracks Factor Level Lower Limit Upper Limit Level 1 > 2 Tracks None Level 3 2 Tracks 2 Tracks Level 5 1 Track 1 Track

7.2.5. Risk Factor 5: Method of Operation

The „Method of Operation‟ risk factor addresses the method by which trains are operated on a subdivision.

This risk factor primarily affects the probability component of risk. Different methods of operation may offer varying levels of reliability/safety in coordinating the movement of trains and in detecting and responding to unsafe conditions that may exist in the operating environment, thereby affecting the frequency with which accidents occur. In general, PTC is expected to provide greater risk reduction on subdivisions where the method of operation currently offers lower levels of reliability/safety. While the implementation of PTC might be expected to substantially decrease the frequency of accidents, and therefore risk, on subdivisions that employ a particular method of operation, it may result in only modest decreases in accident frequency and risk for territories operated under other methods of operation.

CN controls all of their subdivisions using one or more of the three general methods of operation: Track Warrant Control (TWC), Automatic Block Signal (ABS), and Traffic Control System (TCS) (some subdivisions employ more than a single method of operation).

The method of operation was identified for each subdivision. Subdivisions were assigned to one of five factor levels according to the table below. For cases in which more than one method of operation is employed on a single subdivision, the risk factor level was determined by determining the percentage of each method of operation on the subdivision. A subdivision needed to have more than 50% of its route miles as TCS or ABS in order to achieve a lower risk rating classification. There is no ATC/ACS method of operation on CN subdivisions.

PTCIP

Table 18 Methods of Operation Risk Factor Levels Factor Levels for Method of Operation Factor Level Description Level 1 ATC/ACS Level 2 TCS Level 3 Bi-Directional ABS Level 4 Directional ABS Level 5 TWC

7.2.6. Risk Factor 6: Speed of Train Operations

This risk factor provides an indication of the maximum authorized speed with which trains are permitted to travel on a subdivision.

The probability of an accident would likely increase as the maximum authorized speed increases due to the fact that increased speeds require greater stopping distances, and provide train operators with less time to react to a potentially-hazardous condition. Signal overruns are generally more likely when traveling at higher speeds.

Since the energy expended in a train collision or derailment increases with the square of train velocity (kinetic energy = ½*mass*velocity2), speed tends to be a critical factor in determining accident consequences. Increased speed will likely lead to increased damage to equipment, and potentially to increased numbers of injuries and fatalities on-board the train. For trains carrying hazardous materials, traveling at relatively high speeds may increase the likelihood and/or volume of a toxic release, again leading to increased economic costs and casualties.

PTC systems are expected to reduce both the accident rate and the consequences associated with this risk factor, as PTC should be able to prevent the occurrence of many would-be accidents, and in cases where PTC is not able to totally prevent an accident from taking place, it should at least limit the consequences by maintaining train speed under prescribed limits.

The maximum authorized speed was identified for each segment of each subdivision and the maximum speed was used. For segments where the maximum authorized speeds vary depending on train type (passenger, freight, etc.), the maximum authorized speed for any train type was used. Limits for five factor levels, provided in the table below, were assigned taking the range of maximum authorized speed data into account and based on the CN subdivision maximum of 79 MPH.

PTCIP

Table 19 Train Speed Risk Factor Levels Factor Levels for Speed of Train Operations Factor Level Lower Limit Upper Limit Level 1 0 MPH <15 MPH Level 2 15 MPH <30 MPH Level 3 30 MPH <60 MPH Level 4 60 MPH <80 MPH Level 5 80 MPH None

7.2.7. Risk Factor 7: Track Grades

This risk factor addresses the slope of the railroad track, as track grade is an expression of the percentage of a track‟s rise for the length of its run.

Grade would primarily affect the probability component of risk due to the fact that instances of higher grade may increase train momentum to the point that a train could “get away”, preventing braking from stopping the train, or in less extreme cases, the increase in momentum could at least make it more difficult for a moving train to come to a stop within the limit of authority. Increased grade could also act to indirectly increase the consequences of an accident by increasing the speed and kinetic energy of the train, leading to increased damage upon collision.

PTC is expected to reduce the risk associated with instances of higher grade, as it would be expected that PTC will enforce speed restrictions, taking the known grade of a segment of track into account in the braking algorithm. This will allow control of the train to be maintained and will allow the train to be stopped within the limit of authority. Reductions in risk, therefore, would be expected to be greater in subdivisions containing segments of higher grade.

For each subdivision, the maximum grade for any portion of the subdivision was used as a measure of grade for the subdivision. Limits for five factor levels, provided in the table below, were assigned taking the range of grade data into account and based on the CN subdivision maximum of 2%.

Table 20 Track Grade Risk Factor Levels Factor Levels for Track Grades Factor Level Lower Limit Upper Limit Level 1 0.0% <0.5% Level 2 0.5% <1.0% Level 3 1.0% <1.5% Level 4 1.5% <2.0% Level 5 2.0% None

7.2.8. Risk Factor 8: Track Curvature

This risk factor provides an indication of the degree of arc in the railroad track.

PTCIP

Risk of an accident would be expected to increase as curvature increases due to the fact that the centrifugal force to the rail at the point of wheel contact is a function of the severity of the curve, the speed of the vehicle, and the mass of the vehicle. When this force at the point of contact becomes too great, it may overcome the resistance of the track and the train may derail due to rail rollover, the car rolling over, or from the combined transverse force exceeding the limit allowed by rail-flange contact. Instances of increased curvature create an increased potential for an accident because they contribute positively to this increased centrifugal force.

PTC is expected to reduce the risk associated with instances of higher curvature, as PTC will enforce civil speed restrictions, which account for the degree of track curvature. This will help ensure that trains travel through areas of curvature at a safe operating speed, thereby decreasing the likelihood of derailment. Reductions in risk, therefore, would be expected to be increased in subdivisions containing higher numbers of curves and instances of significant curvature.

For each subdivision, the maximum curvature for any section of the subdivision was used as a measure of curvature on the subdivision. Limits for five factor levels, provided in the table below, were assigned taking the range of curvature data into account based on the CN subdivision maximum of 13.95°.

Table 21 Track Curvature Risk Factor Levels Factor Levels for Track Curvature Factor Level Lower Limit Upper Limit Level 1 0.0° <2.0° Level 2 2.0° <4.0° Level 3 4.0° <6.0° Level 4 6.0° <8.0° Level 5 8.0° None

CN also considered whether the additional risk factors identified in the FRA Risk Prioritization Methodology for PTC System Implementation and those in Appendix B should be considered for inclusion in the risk prioritization model. While these other potential sources for risk were discussed, it was estimated that these other factors would have a negligible effect on risk relative to many of the other risk factors that have already been considered in subsections 7.2.1 to 7.2.8 above.

A summary of the risk factor weights used is found in the table below. A detailed discussion of the various approaches evaluated and used for the estimation of risk factor weights is provided in Appendix B.

The outcome was that each risk factor was assigned a qualitative effect on the probability component of risk and on the consequence component of risk; these effects were Very High, High, Medium, Low, Very Low, and Negligible. To aid in quantifying the risk factor weight, each effect was assigned a value in the range of 0 (Negligible) to 1 (Very High). Using this quantification, a sum of the effects on the risk components (probability and consequence) for each risk factor was calculated providing an “effect” score for each risk factor. Lastly, the

PTCIP

ratio of the “effect” score for each risk factor to the sum of the “effect” scores for all risk factors was calculated resulting in the risk factor weights (percentage) in the table below.

Table 22 Risk Factor Weights RF # Risk Factor Weight 1 Annual Million Gross Ton (MGT) 21% 2 Presence and Volume of Passenger Traffic 21% 3 Presence and Volume of TIH/PIH Material 20% (Loads and Residue) Transported 4 Number of Tracks 10.5% 5 Method of Operation 10.5% 6 Speed of Train Operations 11% 7 Track Grades 4% 8 Track Curvature 2%

Details of the risk prioritization model, the risk factors considered, and the quantification of the risk factor weights and levels are provided in the Appendix B.

7.3. Overall Risk Ranking

The risk for each subdivision was calculated using the risk prioritization model described in section 6.2 above. That is, the eight (8) identified risk factors were assigned integer scores, corresponding with level of risk, ranging from 0 (lowest risk) up to 5 (highest risk). Each risk factor was also assigned a weight (percentage) which provides an indication of the “relative importance” of the factor in determining the overall risk ranking. The product of this risk factor level and the risk factor weight produced the weighted effect of the risk factor. Lastly, the sum of the weighted effect of each risk factor was calculated to produce a weighted priority (or relative risk score) of all the risk factors for a subdivision. The weighted priority was then used to determine the risk ranking (risk factor priority) for each subdivision. The following table defines the risk factor priority ranking (integer 1 through 5) based on the relative risk score.

Table 23 Risk Factor Priority Ranking Risk Factor Relative Risk Priority (1-5) Score 0 0 - <1.5 1 1.5 < 2.0 2 2.0 < 2.5 3 2.5 < 3.0 4 3.0 < 3.5 5 > 3.5

The table below summarizes the risk factor priority of each subdivision determined to require PTC. See section 8 for the proposed deployment sequence.

PTCIP

Table 24 Line Segment Risk Ranking Num Subdivision Avg Peak TIH/PIH Passenger Mainline Risk Weighted MGT MGT Cars Trains/Day Route Miles Factor Priority Group 1 Waukesha 52.3 61.6 2,508 22 141.7 5 4.075 2 Chicago 39.7 43.8 8,866 6 121.9 5 3.585 3 Champaign 39.6 51.5 9,344 6 123.1 5 3.565 4 South Bend 42.3 51.9 6,016 8 142.5 4 3.395 5 Yazoo 45.2 48.4 9,099 2 204.1 4 3.375 6 Fulton 40.1 50.2 10,846 2 118.7 4 3.365 7 Centralia 29.4 39.5 4,288 6 115.9 4 3.175 8 Flint 42.4 53.4 8,509 2 155.6 4 3.165 9 Memphis 15.2 49.8 11,305 2 2.5 4 3.110 10 Freeport 9.9 11.8 1,997 16 112.2 4 3.065 11 Neenah 52.9 57.1 2,590 0 86.6 4 3.005 12 Shelby 20.9 24.0 10,846 2 16.6 3 2.990 13 P&I RR 46.6 46.6 2,689 0 13.0 3 2.935 14 Bluford 26.1 46.6 11,035 0 163.6 3 2.885 15 McComb 24.2 31.2 17,852 2 181.4 3 2.885 16 Shore Line 16.2 41.9 2,964 6 51.0 3 2.855 17 Superior 39.8 55.2 2,277 0 233.3 3 2.835 18 Rainy 38.5 44.5 1,451 0 152.2 3 2.795 19 Holly 7.0 9.0 3,399 6 62.9 3 2.775 20 Sprague (US) 44.7 46.0 1,531 0 43.4 3 2.735 21 Mount Clemens 17.3 17.3 6,753 0 45.6 3 2.700 22 Elsdon 26.0 26.0 6,655 2 16.8 3 2.685 23 Baton Rouge 11.5 15.0 22,815 0 79.4 3 2.645 24 Beaumont 11.2 15.6 8,565 0 168.8 3 2.610 25 Cairo 22.4 22.5 4,139 2 41.7 3 2.545 26 Missabe 36.3 74.5 729 0 63.5 3 2.525 27 Joliet 4.3 6.8 4,215 16 33.2 3 2.505 28 Peoria 5.1 8.0 2,247 0 40.1 2 2.390 29 Hammond 5.4 5.4 4,524 0 42.7 2 2.370 30 Leithton 19.3 43.5 2,429 0 65.0 2 2.265 31 Matteson 11.5 44.4 5,971 0 43.4 2 2.255 32 Valley 3.2 5.5 863 0 12.3 2 2.190 33 Gilman 2.4 5.5 589 0 28.9 2 2.155 34 Flat Rock 7.8 13.2 1,447 0 23.9 1 1.995 35 Cherokee 5.7 6.3 29 0 127.3 1 1.950 36 Minneapolis 5.7 5.7 24 0 123.6 1 1.950 37 Fox River 9.1 20.0 27 0 34.9 1 1.930 38 Dubuque 11.6 14.1 649 0 156.9 1 1.895 39 Waterloo 9.7 9.7 463 0 103.8 1 1.875 40 St.Louis 8.3 12.9 29 0 37.1 1 1.725

Risk Factor Groupings - Based on Weighted Average of Risk Factors 0 Weighted priority <1.5 3 Weighted priority from 2.5 to <3.0 1 Weighted priority 1.5 to <2.0 4 Weighted priority from 3.0 to <3.5 2 Weighted priority 2.0 to <2.5 5 Weighted priority 3.5 & over

PTCIP

8. Deployment Sequence and Schedule [§236.1011(a)(5)]

This section details the sequence, planned schedule, and decision criteria for equipping line segments with PTC based on the weighted risk ranking analysis from the previous section. Also included is the proposed schedule for commencing revenue-service PTC operations by December 31, 2015, on all line segments identified as requiring PTC installation.

8.1. CN Key Service Corridors CN‟s track network in the US is generally “Y” shaped with arms extending from Ranier, Minnesota and Port Huron, Michigan that meet in Chicago. The CN network then runs southward from Chicago to New Orleans to complete the “Y”. Each segment of the “Y” is typically a linear network with little option for alternate traffic routing within the CN network.

Subdivisions and tracks comprising each leg of the CN “Y” form a service corridor and our scheduled train service operation is based on the performance of our trains through each of these corridors. This corridor based train service delineation naturally generates a similar alignment for our field engineering and maintenance activities. To support our train service corridors, CN has also aligned our internal data and information systems and our managerial organizational structure along the corridor concept as well.

8.2. CN PTC Corridor Deployment Approach The PTC implementation target is very aggressive and CN wants to ensure that we are able to complete the program within the established schedule by taking advantage of every opportunity possible to improve both the efficiency and effectiveness of the resources allocated to the PTC deployment program as well as the utilization of PTC systems and equipment. One of the ways that this can be achieved is to group main line track segments that require PTC into deployment groupings that align with our existing service corridors or are geographically proximate to the service corridor. This will also assist in using existing systems and data for project metrics and reporting. Prioritization of the deployment groupings will be based on a weighted average of the core main line track segments that form the Service Corridor.

This grouping was done to accommodate a practical approach to the overall system deployment. Rather than attempt a haphazard deployment based solely on subdivision risk ranking, this grouping allows CN to simplify deployment logistics by keeping installation, test, and maintenance crews together as a larger section of the railroad is equipped and PTC is deployed into service. This approach also allows the railroad to take advantage of any PTC benefits sooner since larger, more integrated, sections of the railroad will be equipped and placed into service at a time.

PTCIP

8.3. CN Deployment Groupings CN has developed 5 PTC line segment deployment groupings based on the service corridors and geographically adjacent proximate subdivisions. These deployment groupings are depicted in Figure 2 (Section 1.1.1) and described below.

8.3.1. Pilot Deployment Group: [§236.1011 (a)(4)&(5)] Pursuant to 49CFR§236.1021 and §236.1011 (a)(4)&(5), CN submitted a “Request for Amendment – CN PTC Pilot Territory” of the PTCIP (the “2010 PTCIP”). This request was approved by FRA on February 1, 2012. Specifically, CN has changed the line segments that comprise the Pilot deployment group identified in the 2010 PTCIP. In the original PTCIP, the CN PTC pilot territory extended from Chicago westward and encompassed the main line portions of the Freeport subdivision (west of Munger MP 34.3) the Waterloo and Dubuque subdivisions as well as the eastern most portion of the Cherokee subdivision. CN designates the McComb and Baton Rouge subdivisions as the new Pilot deployment group.

Table 25 Pilot Deployment Group Pilot 2008 Traffic Data Main Line - Over 5MGT or Passenger Traffic Restricted Main Deployment Speed Line Group Avg Peak TIH/PIH Psgr From From To To Route Track (Route MGT MGT Cars Trains MP Station MP Station Miles (miles) Miles) Baton Rouge Baton Rouge 11.5 15.0 22,815 0 364.8 444.2 Orleans Jct 79.4 0.0 79.4 Jct McComb 24.2 31.2 17,852 2 727.2 Jackson 904.4 Mays Yard 177.2 0.0 177.2 Total Route Miles PTC 256.6

8.3.2. Central Deployment Group: Chicago to Memphis The Central PTC Deployment Grouping is based on the CN Service Corridor between Chicago southward to the city of Memphis. This deployment grouping includes some of CN‟s busiest track segments in the US and as well includes the majority of passenger train operations based on train miles.

The core subdivisions that are incorporated in this service corridor are the Chicago, Champaign, Centralia, Cairo and Fulton subdivisions. Geographical proximate subdivisions that have been included in this deployment grouping are the Gilman, Peoria and St. Louis

Table 26 Central Deployment Group Central 2008 Traffic Data Main Line - Over 5MGT or Passenger Traffic Deployment Restricted Main Group Speed Line Avg Peak TIH/PIH Psgr From From To To Route Track (Route MGT MGT Cars Trains MP Station MP Station Miles (miles) Miles)

PTCIP

8.3.3. Gulf Deployment Group: Memphis to New Orleans The Gulf PTC Deployment Grouping is based on the CN Service Corridor between the cities of Memphis and New Orleans. This deployment grouping includes some of CN‟s track segments with the highest TIH traffic volumes and as well includes Amtrak passenger train operations along the entire service corridor.

The core subdivisions that are incorporated in this service corridor are the Memphis, Shelby, and Yazoo. Geographical proximate subdivisions that have been included in this deployment grouping are the Hammond and Beaumont. In addition, this deployment grouping also includes the Bluford subdivision and P&I Railroad. The Bluford and P&I Railroad have been included in this deployment grouping to help balance the workloads between the deployment groupings as well as to minimize potential train service impacts by working on both CN lines north of Memphis at the same time.

Table 27 Gulf Deployment Group Gulf 2008 Traffic Data Main Line - Over 5MGT or Passenger Traffic Deployment Group Restricted Main Speed Line Avg Peak TIH/PIH Psgr From From To To Route Track (Route MGT MGT Cars Trains MP Station MP Station Miles (miles) Miles) Beaumont 11.2 15.6 8,565 0 0.0 Mobile 185.0 Switchtender 185.0 16.2 168.8 Bluford 26.1 46.6 11,035 0 40.7 North Siding 0.0 Edgewood Jct 163.6 0.0 163.6 Hammond 5.4 5.4 4,524 0 43.7 Hammond 0.0 Baton Rouge Jct 43.7 1.0 42.7 Memphis 15.2 49.8 11,305 2 394.3 Y&MV Main 396.8 East Jct 2.5 2.5 2.5 P&I RR 46.6 46.6 2,689 0 0.0 Burlington Jct 14.0 P&I Jct 14.0 1.0 13.0 Shelby 20.9 24.0 10,846 2 390.0 Aulon 13.1 Lakeview 16.6 0.0 16.6 Yazoo 45.2 48.4 9,099 2 13.1 Lakeview 218.6 Jackson 205.5 1.4 204.1 Total Route Miles PTC 611.3

8.3.3.1. Deployment in Eldorado Subdivision [§236.1005 (b)(4)(i)]

Pursuant to 49CFR§236.1021 and §236.1005 (b)(4)(i), CN submitted a “Request for Amendment – Eldorado Subdivision” of the PTCIP (the “2010 PTCIP”). This request was approved by FRA on September 30, 2010. Specifically, CN requested removal of the requirement to install a PTC system on the 29-mile segment of the Eldorado Subdivision identified in the 2010 PTCIP, because less than 5 MGT of traffic has been transported over this segment for two consecutive years, no passenger traffic operates over this segment, and the segment therefore no longer qualifies under the FRA‟s rules as main line track for which a PTC system must be installed.

8.3.4. East Deployment Group: Chicago to Port Huron The East PTC Deployment Grouping is based on the CN Service Corridor between Chicago and the Port Huron tunnel that serves as the primary conduit to CN‟s Eastern Canadian operation. This deployment grouping also includes a number of subdivisions with Amtrak passenger and Metra commuter train operations.

PTCIP

The core subdivisions that are incorporated in this service corridor are the Flint and South Bend subdivisions as well as tracks with significant passenger operations including the Freeport, Eldson, Holly and Joliet subdivisions. Geographical proximate subdivisions that have been included in this deployment grouping are the Cherokee, Dubuque, Waterloo, Flat Rock, Mount Clemens, and Shoreline subdivisions.

Table 28 East Deployment Group East 2008 Traffic Data Main Line - Over 5MGT or Passenger Traffic Deployment Group Restricted Main Speed Line Avg Peak TIH/PIH Psgr From From To To Route Track (Route MGT MGT Cars Trains MP Station MP Station Miles (miles) Miles) Cherokee 5.7 6.3 29 0 381.2 Tara 508.8 Sioux City 127.6 0.3 127.3 Dubuque 11.6 14.1 649 0 115.1 East Jct 272 Hilltop 133.7 0.0 133.7 Elsdon 26.0 26.0 6,655 2 19.3 Blue Island 36.1 Griffith 16.8 0.0 16.8 Flat Rock 7.8 13.2 1,447 0 11.1 FN 39.8 Diann 28.7 4.8 23.9 Flint 42.4 53.4 8,509 2 178.6 Emmett St. 334.2 Port Huron 155.6 0.0 155.6 Freeport 9.9 11.8 1,997 16 2.9 Cermak 115.1 East Jct 112.2 0.0 112.2 Holly 7.0 9.0 3,399 6 4.1 Milwaukee Jct 67.0 Durand 62.9 0.0 62.9 Joliet 4.3 6.8 4,215 16 3.5 Lemoyne 36.7 Jackson Street 33.2 0.0 33.2 Mt. Clemens 17.3 17.3 6,753 0 55.6 Tappan 4.6 Milwaukee Jct 51.0 5.4 45.6 Shore Line 16.2 41.9 2,964 6 54.8 Milwaukee Jct 0 Manhattan Jct 54.8 3.8 51.0 South Bend 42.3 51.9 6,016 8 36.1 Griffith 178.6 Emmett St. 142.5 0.0 142.5 Waterloo 9.7 9.7 463 0 272.0 Hilltop 381.2 Tara 109.2 5.4 103.8 Total Route Miles PTC 1008.5

8.3.5. North Deployment Group: Chicago to Ranier The North PTC Deployment Grouping is based on the CN Service Corridor between Chicago and Ranier which serves as the primary connection to CN‟s Western Canadian operation. This deployment grouping only has passenger train operations on the southern portion of the Waukesha subdivision (MP 15 to 55) and includes the key EJ&E tracks around the western portion of Chicago.

The core subdivisions that are incorporated in this service corridor are the Waukesha, Neenah, Superior, Rainy and Missabe subdivisions as well as the EJ&E Matteson and Leithton subdivisions. Geographical proximate subdivisions that have been included in this deployment grouping are the Fox River, Minneapolis and Valley subdivisions.

In addition to the above, this deployment grouping includes the portion of CN‟s Sprague subdivision that crosses from Canada to the US between Rainy River, Minnesota and International Boundary, Minnesota.

Table 29 North Deployment Group North 2008 Traffic Data Main Line - Over 5MGT or Passenger Traffic Deployment Restricted Main Group Speed Line Avg Peak TIH/PIH Psgr From From To To Route Track (Route MGT MGT Cars Trains MP Station MP Station Miles (miles) Miles) Fox River 9.1 20.0 27 0 208.1 Neenah North 1.4 Green Bay 36.3 1.4 34.9 Leithton 19.3 43.5 2,429 0 65.5 Rondout 101.8 E. Bridge Jct 63.7 0.0 63.7 Matteson 11.5 44.4 5,971 0 101.8 E. Bridge Jct 45.4 Kirk Yard Jct 47.1 3.7 43.4 Minneapolis 5.7 5.7 24 0 308.5 Owen 432.1 Withrow 123.6 0.0 123.6 Missabe 36.3 74.5 729 0 0.5 Duluth Docks 68.6 Largo 68.1 4.6 63.5

PTCIP

North 2008 Traffic Data Main Line - Over 5MGT or Passenger Traffic Deployment Restricted Main Group Speed Line Avg Peak TIH/PIH Psgr From From To To Route Track (Route MGT MGT Cars Trains MP Station MP Station Miles (miles) Miles) Neenah 52.9 57.1 2,590 0 158.4 Shops Yard 247.0 Hoover 88.6 2.0 86.6 Rainy 38.5 44.5 1,451 0 10.7 Nopeming Jct 165.2 Ranier 154.5 2.3 152.2 Sprague (US) 44.7 46.0 1,531 0 1.6 Baudette 45.0 Int. Boundary 43.4 0.0 43.4 Superior 39.8 55.2 2,277 0 247.0 Hoover 480.3 Carson 233.3 0.0 233.3 Valley 3.2 5.5 863 0 63.3 Junction City 49.9 Wisconsin Rapids 13.4 1.1 12.3 Waukesha 52.3 61.6 2,508 21.5 15.5 Tower B12 158.4 Shops Yard 142.9 1.2 141.7 Total Route Miles PTC 998.6

8.4. Deployment Group Weighted Risk Ranking [§236.1011(a)(5)(iii)] RSIA and FRA‟s regulations at 49CFR§236.1011(a)(4) require that PTC be deployed, to the extent practical, in areas of greater risk to the public and railroad employees before areas of lower risk. All of the CN main line track segments that require PTC installation were evaluated using the risk ranking methodology described in section 7 of this document. The risk ranking was performed using the risk factors as required in 49CFR§236.1011(a)(5), to establish risk ratings for each CN subdivision where PTC is required.

CN is using a risk based deployment sequence for staged implementation of PTC on its territories. The analysis provided in Section 7 of this PTCIP is used as the basis for the sequence. As previously discussed, CN has arranged its subdivisions requiring PTC installation into 5 deployment groups and a summary risk ranking for each deployment group of subdivisions was tabulated and used to determine which group would be prioritized first.

The summary risk ranking was calculated based on a weighted average of the line segment weighted risk rankings from Section 7 for all of the subdivisions that formed part of the service corridor or had passenger train operations (excludes the geographically proximate subdivisions which typically have less traffic and no passenger operations). The summary risk ranking values for the 5 deployment groups are as shown below.

1. PTC Pilot Deployment Group –260.8 Route Miles – Summary Risk Ranking 2.88 2. Central Deployment Group – 627.4 Route Miles – Summary Risk Ranking 3.36 3. Gulf Deployment Group – 611.3 Route Miles – Summary Risk Ranking 3.13 4. East Deployment Group – 1008.8 Route Miles – Summary Risk Ranking 2.62 5. North Deployment Group – 999.9 Route Miles – Summary Risk Ranking 2.95

Note: The summary deployment group risk ranking value was calculated using the equation below: n n SummaryRiskRanking  SRiSMi SMi i1 i1

where: n = number of line segments in the deployment group SRi = line segment risk ranking SMi = line segment route miles

PTCIP

As shown by the summary risk rankings above, with the exception of the Central deployment group the risks are essentially equal throughout the CN deployment groups. From this conclusion and other considerations, CN has chosen to install the PTC system starting in a manner that not only aims to reduce safety risk, but also reduces the installation physical risks, the financial risks, and the risks of adverse PTC impact on the normal operations of the railroad. This group deployment strategy is discussed further in Section 8.8.

8.5. Deployment Group Traffic Characteristics [§236.1011(a)(5)(i)] CN traffic characteristics for subdivisions requiring PTC implementation are included in section 6 for main line track segment determination as well as in section 7 for risk prioritization. The table below summarizes these key traffic characteristics by deployment group. A more detailed table that summarizes traffic characteristics by line segment within each deployment group is included in Table_44 of Appendix A.

Table 30 Deployment Group Traffic Characteristics Deployment 2008 Traffic Data Group Weighted Peak Annual Annual Annual Avg MGT TIH/PIH Other Passenger MGT Segment Cars Hazmat Miles Pilot 20.9 31.2 40,667 240,860 132,422 Central 31.7 50.2 40,348 423,919 820,593 Gulf 27.0 49.8 58,063 461,814 165,783 East 22.3 53.4 43,096 912,592 354,436 North 34.4 74.5 20,400 666,946 225,335

From the table above it is clear that the Central deployment group has the highest annual passenger train miles while the Gulf deployment group has the largest annual number of TIH/PIH movements.

8.6. Deployment Group Operational Characteristics [§236.1011(a)(5)(ii)] This section summarizes the operational characteristics by deployment group, including the current method of operation, number of main tracks, and maximum allowable train speeds. A more detailed table of each CN track segment and it‟s operational characteristics is included in Table_45 of Appendix A.

Table 31 Deployment Group Operational Characteristics Deployment Method of Operations Miles of Main Track Max Group Train ABS/ Yard/ Total Total Speed TCS TWC TWC 520 Miles Main1 Main2 Main3 Main4 Miles Pilot 209.9 80.5 0.0 7.3 297.7 260.8 36.9 0.0 0.0 297.7 79.0 Central 654.7 35.3 38.1 3.2 731.3 627.4 90.2 8.1 5.6 731.3 79.0

PTCIP

Gulf 344.1 58.0 211.5 15.7 629.3 611.3 18.0 0.0 0.0 629.3 79.0 East 993.1 121.6 171.1 3.4 1289.2 1008.8 277.3 3.1 0.0 1289.2 79.0 North 781.2 89.4 223.6 0.0 1094.2 999.9 88.8 5.5 0.0 1094.2 60.0

The table above illustrates that while the East deployment group has the largest number of track miles the other deployment groups (excluding the Pilot group) are approximately equivalent in overall track miles and the Gulf group has the largest number of route miles operating under TWC.

8.7. Deployment Group Attributes [§236.1011(a)(5)(iii)] This section provides additional details on CN track route characteristics that may have a bearing on risk. Additional details for most attributes can be found in Appendix A (Line Segment Attributes Detail Tables) or Appendix B (Risk Factor Prioritization Model).

8.7.1. Grade, Curvature, Switches & Road Crossings The table below provides a summary of key attributes on tracks included in the CN PTC deployment plan. Table_46 in Appendix A provides this data for each line segment.

Table 32 Grade, Curvature, Switches & Road Crossings by Deployment Group

Group Track Attribute Road Crossings Switches Max Grade Curves Route Track Total Num/Mile Total Num/Mile (Percent) Total Max(Deg) Miles Miles

Pilot 543 2.1 211 0.7 0.50 216 7.63 260.8 297.7 Central 920 1.5 422 0.6 1.12 485 6.18 627.4 731.3 Gulf 924 1.5 295 0.5 2.00 543 11.49 611.3 629.3 East 1831 1.8 715 0.6 1.40 667 11.93 1008.8 1289.2 North 1326 1.3 688 0.6 1.20 780 13.95 999.9 1094.2 The table above illustrates that all deployment groups are approximately equivalent from the perspective of numbers of switches per mile as well as number of curves per mile, while the East grouping has the highest number of road crossings per mile.

8.7.2. Rail to Rail Crossings at Grade There are a total of 97 railway crossings at grade on tracks included in the CN PTC deployment plan. The table below provides a summary of railway crossings by PTC deployment group. As part of our PTC deployment plan CN will evaluate each of these railway crossings and ensure the PTC system fully meets the requirements for rail-to-rail crossings at grade as defined in 49CFR§236.1005 (a)(1)(i). Table_47 in Appendix A is a more detailed table that provides specific information on each railway crossing.

Table 33 Rail to Rail Crossings at Grade by Deployment Group Deployment Group Railway Crossings at Grade

Pilot 2

PTCIP

Deployment Group Railway Crossings at Grade

Central 22 Gulf 6 East 41 North 24 8.7.3. Movable Bridges There are a total of 7 movable bridges on tracks included in the CN PTC deployment plan. All of the bridges are currently protected with either a stand alone signal system that indicates to approaching trains the alignment status of the bridge or there is interlocking logic connected to the TCS system that prevents the clearing of signals over the bridge unless it is confirmed that the bridge is lined and locked for train operations. As part of our PTC implementation plan CN will ensure that signals governing movements over these bridges are protected by the PTC system.

Table 34 Movable Bridges by Deployment Group Deployment Movable Bridges Group Subdivision Mileage Description Pilot McComb 874.60 Manchac Bridge at Hammond Central None East Mt Clemens 58.20 Black River Bridge at Port Huron Dubuque 182.00 Mississippi River Bridge at Manchester North Rainy 165.44 Rainy River Bridge at Ranier Neenah 173.22 Fox River Bridge at Oshkosh Neenah 209.95 Wolf River Bridge at Gills Landing Leithton 1.70 Des Plaines River Bridge (aka Bridge #198) at Chicago

8.7.4. Passenger Operations All four of the non-pilot CN deployment groups of tracks have regularly scheduled passenger train operations on at least some of their line segments. All passenger operations are trains operated by either Amtrak or Metra. The table below provides a summary of the annual train operations (excluding passenger operations on track with a pending MTEA request). Table_48, Table_49, and Table_50 in Appendix A provide a more detailed summary of annual passenger train operations on CN PTC designated tracks.

Table 35 Annual Passenger Train Operations by Deployment Group

Passenger Train Summary - Annual by Deployment Group Deployment Amtrak Metra Total Group Passenger Total Total Miles (Est.) Total Total Miles Miles Trains Trains (Est.) (Est.) Pilot 730 132,422 0 0 132,422 Central 8,760 820,593 0 0 820,593

PTCIP

Gulf 730 150,015 0 0 150,015 East 16,790 307,768 4,380 55,532 363,300 North 0 0 7,848 225,335 225,335

Note: The data in the table above represents the total train count numbers including all train operations on each line segment within the deployment group. Therefore a single passenger train that operates over three contiguous line segments will be shown in the count above as 3 trains. This methodology skews the actual train count information but presents a more balanced approximation of passenger train operations than counting a train that travels across 5 or 6 line segments for several hundred miles the same as one that travels for 2 miles on a single line segment.

The annual passenger train miles information clearly illustrates that a large percentage (48%) of CN‟s total passenger train miles is in the Central deployment grouping of tracks.

CN has a total of 25 Amtrak and 18 Metra passenger train stations on track that is planned for PTC deployment. These stations are summarized in the table below.

Table 36 Passenger Stations by PTC Deployment Group Deployment Passenger Station Summary Group Amtrak Metra Stations Corridor Stations Corridor

Pilot 1 (McComb subdiv.) None Central 10 Illinois/Saluki None Gulf 3 Illinois/Saluki None East 11 Wolveines/Blue Water 4 Heritage North None 14 North Central Total 25 18 Presence of Other Traffic – Shared Routes There are a number of Class 2 and Class 3 railroads who have running rights to operate trains over CN tracks that are included in the CN PTC deployment plan. The table below provides a summary of the line segments with Class 2 and Class 3 train operation and the typical weekly traffic volumes. This table excludes other Class 1 railroads that have running rights on CN as it is expected that all of these trains will be PTC equipped. The table also excludes Amtrak and Metra train operations as these are specifically identified elsewhere in the PTCIP.

Table 37 Shortline Traffic by Deployment Group Group Subdivision Shortline RailRoad Running Rights Agreements RR MP From MP Trains Description To /Week Pilot No Shortline Operations Central No Shortline Operations Gulf Beaumont MSE 37.8 38.0 4 MSE allowed to use a portion of CN's track to access plant owned by MSE. 2 round trips per week

PTCIP

Yazoo CG 120.6 125.7 6 Use of 5 miles of CN main track. 3 Round trips per week. Yazoo C&J 93.7 ?? 2 Us of CN track to service industries. 1 round trip per week. East Elsdon IHB 8.7 7.6 1 Very few trains operated on this section.

Holly HESR 66.0 69.0 18 HESR uses CN track between Pitt and HESR track at Durand. Holly MDOT - 65.5 69.0 10 Trackage rights for TSBY trains. 5 round TSBY trips per week.

Joliet CSSSB 14.0 11.8 12 Trackage rights. 6 round trips per week. Most movements are on thoroughfare track not mainline. Waterloo IANR 275.1 283.5 4 Trackage Rights. 2 round trips per week North Waukesha WSOR 97.2 122.5 14 Trackage rights Slinger ot Waukesha. 7 round trips per week.

Note: IANR – Iowa Northern Railway MSE – Mississippi Export CG – Central Gulf Railway C&J – C&J Railroad Company (Mississippi Delta Railroad) IHB – Indiana Harbour Belt Railway HESR – Huron and Eastern Railway TSBY – Tuscola and Saginaw Bay Railway Company CSSSB – Chicago South Shore and South Bend Railroad WSOB – Wisconsin and Southern Railway

PTCIP

8.8. Proposed Deployment Schedule The first group of subdivisions chosen for PTC deployment is the Pilot Deployment group consisting of the Baton Rouge and McComb subdivisions. This Pilot Deployment Group addresses FRA‟s objective, as expressed in § 236.1011(a)(4), for PTC systems to be implemented to address areas of greater risk before areas of lesser risk, because it would result in the earlier deployment of PTC on segments that present a relatively greater risk of PTC- preventable incidents. There are high number of TIH car movements on the Baton Rouge and McComb subdivisions. Also, the choice of the Baton Rouge and McCombs subdivisions for the Pilot Deployment Group eliminates a significant risk that CN‟s pilot deployment program might otherwise be disrupted by the removal of line segments that may be determined to be exempt from PTC system implementation.

The second group of subdivisions scheduled for PTC deployment will be the Central. The Central deployment group was selected next as it has the highest overall summary risk ranking of any of the deployment groups and it also includes two of the three highest overall risk ranked line segments (Chicago and Champaign subdivisions). In addition, this subdivision has Amtrak passenger train operations along the entire length of the key service corridor which incorporates 5 subdivisions (Chicago, Champaign, Centralia, Cairo and Fulton).

The third group of subdivisions to have PTC deployed will be the Gulf. The Gulf deployment group was chosen next for PTC implementation to complete the implementation of PTC on the long haul Amtrak corridor from Chicago to New Orleans. An additional benefit of targeting the Gulf deployment group next is to maximize the efficiency of the PTC installation resources as well as to enable maximum utilization of PTC equipped locomotives.

The fourth group of subdivisions that will have PTC deployed will be the East. This group is scheduled next as it contains the majority of the remaining subdivisions with passenger train traffic (Elsdon, Flint, Holly, Joliet, Shoreline and South Bend) and includes more populated areas than the remaining North Group. This deployment group contains a disproportionately large percentage of the interlockings with other roads and requires installation work in some of the more densely congested areas of the CN Network. Scheduling this group later in the deployment will allow CN time to improve installation efficiency and deployment techniques to reduce potential service disruptions to CN, Amtrak, Metra as well other freight carriers.The last deployment group to be progressed will be the North. The North deployment group includes the EJ&E track around Chicago as well as most of CN‟s more isolated or remote tracks in Wisconsin. Scheduling this deployment last enables CN to complete the planned CTC and infrastructure upgrades to the EJ&E tracks in advance of the PTC installation. As part of the planned infrastructure upgrades, CN will be installing wayside equipment and telecommunication facilities to support the planned PTC installation and by scheduling PTC later in the program, we anticipate achieving significant efficiencies when PTC implementation does move ahead. Scheduling this group last also provides ample time for CN to complete our ongoing upgrade of our telecommunications infrastructure in Wisconsin which is critical to providing the required PTC data circuits to support the PTC implementation.

The following figure provides a high level overview of the proposed CN PTC deployment schedule.

PTCIP

Figure 6 CN PTC Deployment Schedule

PTC SubSystem 2010 2011 2012 2013 2014 2015 Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q Q 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 Dispatch Office Dispatch System Upgrade

CN Data System Interfaces

Office Server Installation

Locomotive - OnBoard 2010 - 12 Units 2011 - 22 Units

2012 - 100 Units

2013 - 260 Units

2014 - 300 Units

2015 - 306 Units Wayside - WIUs Pilot - McComb/Baton Rfouge Central - Chicago to Memphis

Gulf - Memphis to New Orleans

East - Chicago to Port Huron

North - Chicago to Ranier

Data Network Build Microwave - Wisconsin

Pilot - McComb Sub

Central - Chicago to Memphis

Gulf - Memphis to New Orleans

East - Chicago to Port Huron North - Chicago to Ranier GIS Design & Build Dbase

GIS Mapping PTC Tracks

GIS Database Maintenance

PTCIP

Exceptions to Risk Based Prioritization [§236.1011 (a)(9)] The final PTC rule Section §236.1011 (a)(4) requires that, to the extent practical, the PTC system be implemented to address areas of greater risk to the public and railroad employees before areas of lesser risk. The following discussion provides details on how the proposed CN PTC deployment plan varies from a pure line segment by line segment risk based PTC implementation.

8.8.1. Corridor Deployment As described earlier in this section, CN is proposing to group PTC required subdivisions into PTC deployment groups that are based on designated service corridors. PTC deployment will be scheduled by deployment group with all subdivisions in one deployment group being completed prior to initiating PTC deployment on the successive deployment group. The benefits of this type of deployment are increased installation efficiency, improved utilization of PTC equipped locomotives as well as reduced service operation disruptions for CN‟s customers. Scheduling of deployment groups is guided by the weighted average risk rankings of the deployment groups. One result of this method of operation is that some lower priority subdivisions in a higher priority grouping may have PTC implemented in advance of a higher priority subdivision in a lower priority deployment grouping.

8.8.2. Geographically Proximate Subdivisions To maximize the efficiency of PTC implementation and minimize the possibility of not completing all required installations by the established deadline of 31 December 2015, CN is proposing to install PTC on some lower priority subdivisions within each deployment group. These subdivisions typically do not form part of the service corridor that the deployment group is based on but are geographically close to the service corridor. CN can improve the implementation efficiency of our PTC initiative by addressing these subdivisions in conjunction with our higher priority service corridor subdivisions.

8.8.3. Subdivisions with Limited Segments of Passenger Operations The final PTC deployment group (North) includes CN‟s Waukesha Subdivision which has the single highest weighted priority ranking of all of the 41 CN subdivisions that require PTC. Based on the high priority assigned to this sub, a further explanation of why it is proposed for later PTC implementation is required. The high priority assigned to the Waukesha subdivision is a result of the high number (22) of daily Metra commuter trains that operate on it. A closer look at the passenger train operations shows that the commuter trains only operate from MP 15.5 to MP 57.2 a total of 41.7 miles. Therefore there are passenger train operations only on approximately 1/3 of the Waukesha subdivision but the passenger train presence impacted the weighted risk ranking as if the passenger trains operated over the entire subdivision. This has the impact of artificially raising the weighted risk ranking of the Waukesha subdivision. In addition, with the purchase of the EJ&E, CN is moving much of its freight traffic off of the south end of the Waukesha subdivision (MP 37.2 to MP 15.5) onto the EJ&E tracks. This change in traffic routing results in freight and passenger operations coinciding for only approximately 20 miles on the subdivision (MP 37.2 to MP 55.7) which significantly reduces relative risk of a PTC preventable incident.

PTCIP

8.9. De-Minimis Exception Requests [§236.1005 (b)(4)(ii)] CN has several track segments that have low freight volumes and minimal volumes of TIH/PIH traffic. These track segments are identified below and are included in section 14 of this document as De Minimis PTC exclusion requests.

The following list provides an overview of the De-Minimis exclusions being requested by CN:

1. Cherokee Subdivision – MP 381.2 to MP 508.8 – 127.6 track miles, 2008 traffic was 5.7 MGT with 29 carloads of TIH/PIH traffic. 2. Minneapolis Subdivision – MP 308.5 to MP 432.1 – 123.6 track miles, 2008 traffic was 5.7 MGT with 24 carloads of TIH/PIH traffic,

CN has fully included the above subdivisions in its PTCIP line segment ranking and risk evaluation criteria. These subdivisions are currently included in the CN PTCIP deployment plan and will be maintained as part of CN‟s PTC deployment plan until such time as approval of the De Minimis exclusion request may be received from the FRA.

Details on all De Minimis exclusions are included in Section 14 of this document.

8.10. Sprague Subdivision – Description and Overview

Sprague Subdivision has a number of unique characteristics that would cause PTC implementation to have significant commercial, operational and economic impacts on CN. These unique characteristics include:

a) Dispatch Office – The Sprague Subdivision is currently controlled from CN‟s dispatch office located in Edmonton, Alberta, Canada. Implementation of PTC on the Sprague Subdivision will require installation of PTC office computers in Edmonton as well as development of a special interface to CN‟s Siemens-developed dispatch system that is unique to Canada.

b) Employee Training – Train and engine crews that operate on the Sprague Subdivision are based out of Winnipeg, Manitoba and Fort Frances/Rainy River, Ontario. Installation of PTC on the Sprague Subdivision will require PTC training and familiarization for a significant number of Canadian based crews that would use the system only irregularly at best. In addition, CN would have to provide training to dispatch office staff responsible for PTC office equipment maintenance.

PTCIP

c) Operating Rules – Trains operating on the Sprague Subdivision currently use operating rules as defined under CROR (Canadian Rail Operating Rules). Installation of PTC will require that the PTC office computer system be re-developed and verified to work with operating rules as defined by CROR.

The portion of the Sprague Subdivision within the United States has been fully included in CN‟s PTC line segment assessment, risk ranking prioritization, and PTC deployment scheduling. CN intends to implement its PTC deployment plan in a manner that fully includes the Sprague Subdivision as required by the RSIA and 49CFR§236; however due to the unique characteristics of the Sprague Subdivision and the potential impacts associated with PTC implementation, CN intends to develop and submit a request for waiver or exemption seeking relief from PTC implementation for the US portion of the Sprague Subdivision.

The following text provides more details on the Sprague Subdivision and outlines a number of potential mitigation techniques that could be considered as alternatives to PTC implementation in order to achieve a similar or reduced level of overall risk.The US portion of the Sprague Subdivision is located in a relatively remote section of northern Minnesota with no major population centers. Although there is no passenger traffic on this line segment, it does have an average annual freight volume of 44.7 MGT and an annual TIH car volume of 1,531 (2008 values). The following is a complete summary of the key attributes of this line segment:

1. Track Miles: MP 1.6 to MP 45.0 – 43.6 miles, 2. Traffic Volume: 2008 – 44.7 MGT, 2009 37.5 MGT, 3. TIH Traffic Volume: 2008 - 1,531 (675L, 856R), 2009 – 1,263 (557L, 706R), 4. Track Class: Class 4 maximum speed 60 mph, 5. Ruling Grade: 0.50% ascending eastbound grade, 6. Method of Operations: 43.6 miles of TCS, 7. Curvature: 8 curves, max. curvature 2.19 degrees, avg 0.18 curves/mile, 8. Crossings: 66 crossings, avg 1.5 crossings/mile, 9. Turnouts: 16 mainline turnouts, 0.4 turnouts/mile, 10. Summary weighted risk ranking: 2.735 ranked 20th out of 41 CN subdivisions that require PTC installation.

Trains operating on this section of track are controlled from CN‟s Edmonton dispatch office with TCS deployed along the entire 43.6 miles. This line segment is relatively straight with only 8 curves and has a flat topography with a maximum grade of 0.5%. There are only 16 mainline switches on this segment of track and there is good radio coverage so trains always have reliable communications to contact the dispatchers as required.

There are a total of 5 sidings on the section of the Sprague Subdivision in the US. Two of the sidings are longer sidings and are the ones that typically are used for train meets. The siding at

PTCIP

Graceton is the only siding that is signaled with track circuits and broken rail protection on the siding track. Siding locations and lengths are:

 Baudette – MP 1.6 – Length 6,857 feet  Graceton – MP 11.3 – Length 10,117 feet  Williams – MP 17.8 – Length 6,700 feet  Blueberry – MP 22.9 – Length 10,303 feet  Swift – MP 31.7 – 6,674 feet

The Sprague Subdivision has the following wayside detection and inspection systems in place to reduce the risk of derailment:

 Wayside Inspection Systems – CN has hot box and dragging equipment detectors located approximately every 15 miles on the Sprague Subdivision. Sites in the US are located at Pitt (MP 14.47), Roosevelt (MP 28.2) and Longworth (MP 42.2). All sites are equipped with hot wheel, hot bearing and dragging equipment detection capability. Sites are equipped to notify trains by radio of any alarm conditions and also report wheel and bearing temperature information to the central CN wayside alarm monitoring system for temperature trending analysis. All alarm conditions are reported to the responsible dispatcher as well as directly to the train crews.

 Wayside Impact Load Detector – CN has a comprehensive system of WILD detection systems to monitor and do trending analysis on wheel impact conditions. One of the WILD detectors is located on the Sprague Subdivision just south of Winnipeg approximately 100 miles from the US border. Data from the WILD detector is fed to CN‟s central WILD monitoring system for trending analysis and automated alarming of high impact conditions to the responsible dispatcher.

There are no major population centers along this segment of track with the largest towns being Warroad, Minnesota (2000 population of 1,722) and Baudette, Minnesota (2000 population of 1,104). Since this area is remote, with no passenger traffic and complete CTC coverage, the primary risk relating to a PTC preventable incident is the potential environmental impact of substantial TIH releases. The greatest risks of TIH hazards are generally attributed to releases due to derailments. Rather than deploy PTC on this isolated subsection, CN will propose to achieve similar or greater levels of risk reduction through alternative means specifically targeted at reducing the chances of derailment. In addition to the existing wayside inspection system coverage and risk reduction measures already in place, CN will propose infrastructure improvements to reduce overall risks which will include some or all of the following:

PTCIP

1. Add an additional WILD detector to monitor northbound train movements over the US portion of the Sprague Subdivision, 2. Upgrade the sidings on the Sprague Subdivision to include track circuit protection to reduce the possibility of broken rail related and other derailments. 3. All CN train movements on the Sprague Subdivision are already operated in full compliance with the requirements of AAR Circular No. OT-55 (Recommended Railroad Operating Practices for Transportation of Hazardous Materials). CN will maintain its compliance with these voluntary recommended operating practices. Since the Positive Train Control Rule promulgated in 49CFRPart 236 I does not appear to authorize an exemption request that fits the criteria of this situation, CN has included the Sprague subdivision in the remainder of this PTCIP but based on the unique facts related to the subdivision, CN expects to submit a waiver or exemption request under general FRA regulations (i.e., outside the PTCIP process) to allow CN to deploy alternate risk reduction techniques in lieu of PTC on this subdivision.

A map of the Sprague Subdivision is attached below (US track highlighted):

Figure 7 Sprague Subdivision

PTCIP

9. Rolling Stock [§236.1011(a)(6)] This section contains information related to the CN rolling stock that will be equipped with the PTC technology.

9.1. CN Locomotive Fleet Overview The CN locomotive fleet consists of a total of 1732 locomotives built by either General Electric Transportation (Erie, Pennsylvania) or Electromotive Diesels (LaGrange, Illinois). All CN locomotives utilize DC traction motor technology and are classified as either Low Horsepower (LHP - under 3,000 horsepower) or High Horsepower (HHP - 3,000 horsepower and greater). Low horsepower locomotives are normally assigned to designated yard or terminal operations while the high horsepower locomotive fleet is not normally assigned to any designated service. The active locomotive fleet at any given time will vary depending on the current traffic levels and operating service plan requirements.

In developing its PTC Implementation Plan, CN reviewed its overall locomotive fleet assignments as well as our operating service plan to determine the best strategy for meeting its PTC objectives. The option of segmenting our HHP locomotive fleet into Canadian and US assignments to reduce PTC implementation costs was considered but rejected in favour of a more aggressive plan to equip the majority of the HHP locomotives. This plan will ensure that trains operating across the border from Canada will have PTC equipped locomotives when they enter the United States.

The table below provides a snapshot of the CN locomotive fleet at the time of submission of this PTC Implementation Plan.

Table 38 CN Locomotive Fleet Horse Power No. of Units HHP 1210 LHP 522 Total 1732

9.2. Locomotives to be PTC Equipped [§236.1011(a)(6)(i)] As part of the PTC implementation initiative, CN will equip 820 of our HHP locomotives with PTC as well as 180 US assigned yard and road switcher locomotives by 31 December 2015. The table below provides the summary of locomotives that will have PTC equipment installed:

Table 39 PTC Equipped Locomotives Model HP Inv Tot PTC Eqp B398 HHP 11 0 C408 (DASH-8) HHP 93 86 C449 (DASH-9) HHP 244 231 ES44DC HHP 90 90 E9A LHP 4 0 F40PH HHP 3 0

PTCIP

Model HP Inv Tot PTC Eqp GMD1 LHP 26 0 GP382 LHP 245 150 GP40 HHP 17 0 GP402 HHP 63 0 GP40R HHP 16 0 GP9RB LHP 20 0 GP9RM LHP 150 28 RS18 LHP 2 0 SD38 LHP 31 2 SD40 HHP 11 0 SD402 HHP 154 20 SD403 HHP 61 24 SD40A HHP 9 0 SD40Q HHP 26 0 SD60 HHP 62 0 SD70 HHP 36 36 SD70I HHP 26 26 SD70M-2 HHP 115 140 SD75I HHP 173 167 SW1001 LHP 1 0 SW1200 LHP 6 0 SW1500 LHP 15 0 SW7RM LHP 5 0 SW7SW9 LHP 17 0 TOTAL 1732 1000

Once all PTC implementations are completed, CN will have equipped over 98% of its total HHP mainline freight road haul locomotives and 85 % of its LHP fleet assigned to US operations. This will provide adequate PTC equipped locomotives to support all CN freight and work train operations as well as local and yard switching operations on PTC equipped tracks.

9.3. Rolling Stock PTC Implementation Schedule [§236.1011(a)(6)(ii)] Pursuant to 49CFR§§236.1021, 236.1005(a)(6)(ii), and 236.1006(b)(1), CN submitted a “Request for Amendment – Rolling Stock and Annual PTC Goals” of the PTCIP. This request was approved by FRA on February 1, 2012. Specifically, CN has (1) amended the scheduled installation of PTC equipment on locomotives (due to delays in the availability of onboard PTC equipment), and (2) included (in Section 9.3.)1 annual goals for PTC-equipped train operations (which were inadvertently omitted from the 2010 PTCIP).

As reported in CN‟s annual update, and as FRA is independently aware, in the interval since CN submitted the 2010 PTCIP the PTC development initiative has encountered extended unanticipated delays in the availability of several key components of PTC onboard equipment, including the messaging processor, messaging software, and the 220 MHz Data Radio. These delays are directly impacting CN‟s ability to meet the onboard PTC equipment installation schedule set forth in Table 40 of the 2010 PTCIP. CN therefore sought FRA approval to

PTCIP amend the schedule. The change will reduce the number of PTC-equipped locomotives scheduled to be deployed in 2011 and 2012, with a corresponding increase in future years so that all 1,000 locomotives identified in Section 9.2 will be PTC-equipped as required by December 31, 2015.

The adjustment to the onboard installation schedule is driven by the unavailability of the equipment, a development outside CN‟s control that has impacted the entire industry. Although this unanticipated development has prevented CN from adhering to the schedule in the 2010 PTCIP, CN is committed to dedicating the resources required to complete deployment on all 1,000 locomotives identified in the PTCIP by December 31, 2015. Based on its understanding of the current anticipated timetable for commercial availability of the necessary equipment, CN believes the December 31, 2015 deadline set by the Rail Safety Improvements Act of 2008 (“RSIA”) remains attainable. The schedule adjustments proposed herein will also enable CN to complete PTC equipment installation on each locomotive in a single session, which will reduce PTC implementation costs and optimize the utilization of resources needed to support the aggressive PTC implementation timeline, as well as eliminate any other potential multi-stage integration issues.

In 2010, a pilot program equipped six (6) EMD and six (6) GE High Horsepower locomotives with PTC equipment. The purpose was to assess and evaluate the optimum equipment layout and installation procedure on two of our key types of HHP road locomotives.

Rollout of further PTC installation on the CN HHP locomotive fleet is scheduled for 2011 with 22 locomotives. Installation then ramps up in 2012, 2013, 2014. The remainder of the HHP locomotives will be completed in 2015. Table 40 below provides more details on the overall locomotive implementation plan.

Installation of PTC equipment on Low Horsepower locomotives will be completed in four (4) groups, with the first group scheduled for conversion in 2013 and successive groups scheduled in 2014 and 2015. Equipping of LHP locomotives has been minimized in the first years of the rollout to help ensure availability of resources to focus priority in 2011 and 2012 on equipping the HHP fleet.

The schedule for installation of PTC onboard equipment is as follows:

Table 40 PTC Onboard Installation Schedule and % Completion Year HHP LHP Total % HHP %LHP Total % 2010 12 0 12 1% 0% 1% 2011 22 0 22 4% 0% 3% 2012 100 0 100 16% 0% 13% 2013 200 60 260 41% 33% 39% 2014 240 60 300 70% 67% 69% 2015 246 60 306 100% 100% 100% Total 820 180 1000

Whenever possible, locomotive PTC modifications will be performed when a locomotive is in one of CN‟s main running repair shops for major repairs, overhauls or quadrennial inspections.

PTCIP

Aligning PTC equipment installation in conjunction with these other activities will assist in scheduling PTC installation activities and help ensure that service impacts are minimized while locomotives are out of service for upgrades.

Installation of PTC equipment on CN locomotives will be performed by CN employees whenever possible. Technical assistance and guidance will be provided by technical service personnel from the PTC equipment suppliers or other industry technical resources as required. In the event the internal CN resources are inadequate to maintain the forecast PTC equipment implementation schedule, CN will contract equipment installation activities to external contract facilities as required. Prior to contracting PTC equipment installation activities, CN will undertake a review and assessment of the technical expertise and ability of the external facility to undertake the PTC installation workload.

As locomotives attrite from the CN fleet, they will be removed from the PTC implementation plan and other locomotives added to maintain the overall targeted number of PTC equipped locomotives. In the future, if locomotive manufacturers become capable of supplying new locomotives that are PTC equipped and enabled from the factory, CN will pursue this option for new locomotive purchases as a component of our strategy for achieving locomotive PTC implementation targets.

In accordance with rule §236.1006(b)(2) CN will report its progress toward achieving its planned PTC locomotive deployment by April 16, of years 2011, 2012, 2013, and 2014.

9.3.1. Establishment of Annual Goals for PTC-Equipped Train Operations To assure technical compliance with 49CFR§236.1006(b)(1), CN has amended the 2010 PTCIP to include annual goals for incremental growth in the percentage of controlling locomotives operating on PTC lines that are equipped with operative PTC equipment. These goals were established using the methodology discussed below.

Using 2010 actual train data, CN determined the average number of trains operated on a daily basis (“train operations”) on line segments scheduled to be equipped with PTC. CN used this number as the theoretical maximum number of opportunities for a train to be operated over PTC- equipped track by a locomotive equipped with operational PTC equipment. This theoretical maximum number was then multiplied by the number of locomotives scheduled to be equipped with PTC in a given year (see Proposed Onboard Installation Schedule provided above), in order to calculate an annual target percentage or goal of train operations that could be made with a locomotive installed with operational PTC equipment on PTC-equipped track.

The following calculation for the 2013 goal, as given in our RFA on Rolling Stock, demonstrates the methodology used to determine the annual goals:

1. In 2010, there were a total of 285,244 discrete train movements on CN line segments located in the U.S. Of that number (285,244), 226,264 involved movements over line segments scheduled for PTC installation. Therefore, the daily average maximum number for PTC-equipped train operations is 620 (226,264/365).

2. By the end of 2012, CN expects to have equipped 134 locomotives with PTC equipment; CN expects to equip an additional 200 locomotives during 2013. For the purpose of setting incremental annual goals under §236.1006(b)(1), it is assumed that the

PTCIP

installations will progress at a constant pace throughout the year. Therefore, CN has assumed that 234 PTC-equipped locomotives will be available on a daily average basis during 2013 (134 completed by year end 2012 plus 100 (50% of the locomotives scheduled for installation in 2013)). As reported in the 2010 PTCIP, a total of 1000 locomotives in all are scheduled for PTC installation, meaning that 23% of the CN locomotive fleet (234 of 1000) will be equipped with PTC on a daily average basis in 2013.

3. By the end of 2012, CN will have completed PTC installation on the two line segments that CN has designated as its Pilot territory (McComb and Baton Rouge) . These line segments have a combined total average of 23.7 train operations per day. (The separate line segments scheduled for PTC installations in 2013 will not be placed into service until late in 2013, and therefore have not been included in determining the annual target for PTC-equipped train operations in 2013.)

4. Based on the calculations and assumptions presented above, the theoretical maximum number of daily PTC-equipped controlling locomotives operating on PTC-equipped lines in 2013 is 5.4 (the average of 23.7 train operations on PTC-equipped track multiplied by 23% (the probability on a daily average basis in 2013 that the controlling locomotive will be PTC-equipped)).

5. Therefore, the maximum potential percentage of PTC-equipped controlling locomotives operating on PTC-equipped lines in 2013 would be 5.4 out of 620, or 0.9%.

Based on goals established using this methodology, CN has amended the 2010 PTCIP by adding the following table of goals in a new subsection to Section 9 of the PTCIP:

Table 41 Proposed Goals for PTC-Equipped Train Operations (as of January 1 of Listed Year) 2010 2011 2012 2013 2014 2015 2016 0% 0% 0% 1.1% 12.9% 54.1% 100%

As of December 31, 2015 CN will have completed the installation of PTC equipment on all 1000 locomotives and on all required line segments identified in the PTCIP (as it may be amended from time to time), enabling all trains operating on PTC track to be deployed with a PTC-equipped locomotive with operational PTC equipment.

9.4. Tenant Railroads [§236.1011(a)(6)(iv)(A) and (B)] Tenant railroads operating on CN track include most of the Class 1 freight railroads, Amtrak and Metra, and the Class 2 and Class 3 railroads that have been identified in section 5 of this document.

For the purpose of this PTCIP submission, CN has signed Interoperability Agreement letters with Amtrak and Metra as well as all of the other Class 1 railways. CN is also working with all the Class 1 railroads through the ITC process. All primary tenants, that is, Class 1 freight railroads, Amtrak and Metra are submitting PTCIP documents independently. Thus, in

PTCIP accordance with rule §236.1011(a)(6)(iv), CN will not be filing the details of their rolling stock nor their deployment plans in its own submission.

Class 2 and Class 3 tenant railroads are not required to be PTC-equipped by rule §236.1006(b)(i-iii), and therefore, there is no deployment information on these tenant railroads available. CN will work closely with all Class 2 and Class 3 tenant railroads to ensure they are informed of our PTC implementation plans.

PTCIP

10. Wayside Devices [§236.1011(a)(7)] As described in section 1.1, the CN PTC system is a locomotive-centric train control system that uses a combination of locomotive, office, and wayside data integrated via a radio network. This section identifies the wayside devices and subsystems that will be installed as part of the CN PTC System.

10.1. Wayside Device Equipment The two major wayside devices are the Wayside Interface Unit (WIU) and the Wayside/Base Communication Packages (BCP).

10.1.1. Wayside Interface Units The WIU is a vital device that reads the real time status of specific interlocking devices(signals and switches), creates pre-programmed messages derived from this data, and transmits this data from the WIU to the locomotive and/or office subsystems of the PTC system using the wayside/base communication network.

Depending on the method of train control (CTC, Track Warrant ABS or Track Warrant), CN is to install ITC specification compliant WIUs at wayside locations as shown in Table 42.

Table 42 WIU Installations

Method of Control CTC Track Warrant ABS Track Warrant Control Points Yes Intermediate & Approach Signals Yes Yes Yes Entering Signals Yes Yes Interlockings Yes Yes Yes Moveable Bridges Yes Yes Yes Hand Throw Switches Yes

The table of Wayside Device Tabulations in section 10.2 provides an estimate of the anticipated WIUs per subdivision.

CN anticipates the use of both stand-alone and integrated WIU platforms.

Stand alone WIUs are designed to monitor signal devices directly and are therefore well suited to installation at hand throw switches in track warrant territory, relay controlled signals, control points and interlockings, and at locations with electronic control equipment that cannot be upgraded to provide WIU functionality. CN expects that this equipment will be employed at approximately 55% of WIU locations.

Integrated WIU platforms are designed as an extension of existing electronic control equipment, commonly adding WIU functionality through an upgraded CPU card. In this case, the control equipment monitors the state of the signal devices and passes this information vitally to the portion of the CPU card implementing the WIU functionality. CN expects that this equipment will be employed at approximately 45% of WIU locations.

PTCIP

10.1.2. Wayside/Base Communications Packages The wayside data communications (or base communication packages) will be via a 220MHz narrowband wireless network. It is the industry-standard private radio implementation, specified and designed by the ITC consortium. The private 220MHz network will support communications between the office, locomotive, and wayside subsystems and will utilize spectrum owned and managed by the ITC consortium. The base communication packages will be comprised of the following major components, each performing one of the BCP‟s primary functions: RF transceiver, RF transmission interface, wireline interface, radio interface and antenna system. These BCPs will be located across the subdivisions as required by the design; the table of Wayside Device Tabulations in section 10.2 provides an estimate of the anticipated BCPs per subdivision.

10.2. Wayside Device Tabulations The following table provides a tabulation of the projected number of major wayside devices (WIUs and BCPs) to be installed by subdivision.

Table 43 Wayside Device Tabulations From From To To Deployment Group Subdivision Name MP Station MP Station # WIUs # BCPs BATON ROUGE 364.8 Baton Rouge Jct 444.2 Orleans Jct 47 4 PILOT MCCOMB 727.2 Jackson 908.6 Mays Yard 85 11 CAIRO 363.1 Illinois 405.4 Cairo Jct 24 2 CENTRALIA 247.2 Sandoval Jct 363.1 Illinois 57 5 CHAMPAIGN 124.1 Leverett Jct 247.2 Sandoval Jct 50 5 CHICAGO 1.5 16th Street 124.1 Leverett Jct 59 9 CENTRAL FULTON 269.2 Cairo Jct 387.9 Leewood 58 9 GILMAN 81.1 Gilman 110.0 Gibson City 54 3 PEORIA 119.2 Mattoon 79.1 Decatur Jct 27 3 ST LOUIS 70.0 Duquoin 32.9 Lenzburg 30 3 BEAUMONT 0.0 Mobile 185.0 Switchtender 45 9 BLUFORD 40.7 North Siding 0.0 Edgewood Jct 40 5 HAMMOND 43.7 Hammond 0.0 Baton Rouge Jct 36 1 GULF MEMPHIS 380.4 Memphis 396.8 East Jct 6 1 P&I RR 0.0 Burlington Jct 14.0 P&I Jct 8 1 SHELBY 387.9 Leewood 13.1 Lakeview 17 0 YAZOO 13.1 West Jct. 218.6 Jackson 95 11 CHEROKEE 381.2 Tara 508.8 Sioux City 50 5 DUBUQUE 115.1 East Jct 272.0 Hilltop 66 8 ELSDON 19.3 Blue Island 36.1 Griffith 21 1 FLAT ROCK 11.1 FN 39.8 Diann 21 3 FLINT 178.6 Emmett St. 334.2 Port Huron 77 5 FREEPORT 2.7 21st Street 115.1 East Jct 70 5 EAST HOLLY 4.1 Milwaukee Jct 67.0 Durand 37 2 JOLIET 3.5 Lemoyne 36.7 Jackson Street 21 3 MT CLEMENS 55.6 Tappan 4.6 Milwaukee Jct 33 1 SHORE LINE 54.8 Milwaukee Jct 0.0 Manhattan Jct 37 1 SOUTH BEND 36.1 Griffith 178.6 Emmett St. 75 3 WATERLOO 272.0 Hilltop 381.2 Tara 59 4

PTCIP

From From To To Deployment Group Subdivision Name MP Station MP Station # WIUs # BCPs FOX RIVER 208.1 Neenah North 1.4 Green Bay 43 4 LEITHTON 65.5 Rondout 101.8 E. Bridge Jct 35 7 MATTESON 101.8 E. Bridge Jct 45.4 Kirk Yard Jct 28 7 MINNEAPOLIS 308.5 Owen 432.1 Withrow 49 6 MISSABE 0.5 Duluth Docks 68.6 Largo 36 1 NORTH NEENAH 158.4 Shops Yard 247.0 Hoover 66 4 RAINY 10.7 Nopeming Jct 165.2 Ranier 72 9 SPRAGUE (US) 1.6 Baudette 45.0 Int. Boundary 59 2 SUPERIOR 247.0 Hoover 480.3 Carson 130 7 VALLEY 63.3 Junction City 49.9 Wisconsin Rapids 31 3 WAUKESHA 15.5 B12 Tower 158.4 Shops Yard 117 8

Note 1) Within the Eldorado Subdivision, PTC implementation will not occur between Eldorado Junction and Akin Junction. See Section 8.3.3.1.

10.3. Wayside Deployment Schedule [§236.1011(a)(5)&(7)] Pursuant to 49CFR§§236.1021, 236.1011(a)(5), and 236.1011(a)(7), CN submitted a “Request for Amendment – Wayside Deployment Schedule” of the PTCIP. This request was approved by FRA on February 1, 2012. Specifically, CN has adjusted the scheduled installation of PTC wayside equipment due to delays in the availability of necessary equipment and software.

As reported in CN‟s annual update, and as FRA is aware independently, since CN submitted the 2010 PTCIP, the PTC development initiative has encountered extended unanticipated delays in the availability of production quantities of key PTC hardware components (primarily the 220 MHz Data Radio required for wayside base stations and wayside WIU installations), the Wayside Messaging Server (WMS), and two key software components of the PTC solution (Messaging Protocol Management software and System Management System software). These delays are directly impacting CN‟s ability to meet the wayside PTC equipment deployment schedule set forth in Figure 6 and Table 43 of the 2010 PTCIP. CN has therefore sought FRA approval to amend the schedule. These changes will reduce the number of PTC-equipped wayside units scheduled to be deployed in 2011 and 2012, with a corresponding increase in wayside deployment in future years so that all line segments that must be PTC-equipped will be completed as required by December 31, 2015.

The schedule changes will not change the number or type of PTC-equipped wayside units that will be installed or the CN line segments that will be PTC-equipped; nor will they alter the line segment deployment sequence that CN presented in the 2010 PTCIP after factoring in relative risk rankings as required by the Rail Safety Improvements Act of 2008 (“RSIA”). The proposed adjustments are driven by the unavailability of PTC equipment, a development outside CN‟s control that has impacted the entire industry. Although this unanticipated development will prevent CN from adhering to the schedule in the 2010 PTCIP, CN is committed to dedicating the resources required to complete PTC wayside deployment by the statutory deadline on all line segments that must be PTC-equipped. Based on its understanding of the current anticipated timetable for commercial availability of the necessary equipment, CN believes the December 31, 2015 deadline set by RSIA remains attainable.

PTCIP

PTC installation is scheduled to start in 2011, with the completion of the various deployment groups, outlined in section 8, scheduled as follows:

Table 44 Percentage of WIUs and BCPs Installed

Deployment Group % of WIUs Installed % of BCPs Installed Scheduled Completion Pilot 7 8 31 December, 2012 Central 28 33 31 December, 2013 Gulf 41 48 31 December, 2014 East 66 68 31 December, 2014 North 100 100 31 December, 2015

Installation of wayside PTC equipment is to be performed by CN employees whenever possible. Technical assistance and guidance will be obtained from field support personnel from the PTC equipment suppliers and other technical resources as required. In the event that internal CN resources are inadequate to maintain the forecast PTC equipment implementation schedule, CN will contract equipment installation activities to external firms as required. Prior to contracting PTC installation activities, CN will undertake a review and assessment of the technical expertise and ability of the external firm to undertake the PTC installation workload.

In accordance with rule §236.1006(b)(2) CN will report its progress toward achieving its planned PTC WIU deployment by April 16, of years 2011, 2012, 2013, and 2014.

PTCIP

11. Submittal Dates for PTCDP and PTCSP [§236.1011(a)(10)] Based on the PTC deployment plan (see section 8) CN anticipates using an unmodified I-ETMS Type Approval (per §236.1009(b)(1)) for its PTC system. On March 24, 2010, UP, NS, and CSXT submitted a PTCDP for the I-ETMS platform. A revised PTCDP, Revision 2.0 was submitted by the same railroads dated June 1, 2011, and a Type Approval was received from FRA. The platform described in that PTCDP is identical to the platform that CN intends to use.

PTC Implementation Plan April 16, 2010 PTC Development PlanType Approval rec‟d August 26, 2011 PTC Safety Plan Definition Document* September 2010 PTC Safety Plan end of June 2013** RFA to either PTCIP***, PTCDP, or PTCSP As appropriate per Rule

* This document (PTCSPDD) describes the intended organization and content of the CN PTCSP document. The PTCSPDD explains how the Safety Program for CN PTC will, in particular, approach the PTCSP requirements covered under FRA Part 236H and Part 236I. This can be used to provide the FRA with advanced knowledge of CN‟s planned approach to each referenced Part 236I rule paragraph, and to potentially obtain FRA feedback as to the correctness of the interpretation of the regulation.

** This is an approximate date based on the current deployment schedule found in Figure 6. It corresponds to the completion of the safety server installation, which is the final element required to certify the Pilot Group deployment. Should there be a further change to this deliverable date, it will be included in a future RFA.

*** As required by the final Rule under §236.1011(f), The PTCIP will be maintained to reflect CN‟s most recent PTC deployment plans until all PTC system deployments required under Subpart I are complete.

PTCIP

12. Strategy for Full PTC System Deployment [§236.1011(b)] CN has identified all subdivisions that require PTC deployment to comply with 49CFRpart 236, subpart I and plans to equip them with PTC during the primary implementation period, per the schedule found in section 8.

CN‟s strategy for full PTC deployment is to evaluate the economic and safety benefits derived from the deployment of PTC on the required subdivisions before making any long term plans beyond the subdivisions required by the rule. Criteria similar to the risk priority parameters stated in Section 7 of this PTCIP may be used in the future to determine if additional elective PTC deployment will be undertaken. Given the extensive effort to equip the mandated subdivisions, such evaluation will be deferred until after the primary implementation period.

CN will also review subdivision traffic patterns as part of its Risk Reduction Program on an annual basis to determine if additional PTC deployment becomes required under the rule on any subdivision.

100

PTCIP

13. Main Line Track Exclusion Addendum [§236.1019]

13.1. MTEA General This Mainline Track Exclusion Addendum seeks to have designated as not “main line” 6 segments of CN‟s track used for scheduled intercity passenger service. These requests are made pursuant to and would be subject to the conditions set forth in 49CFR§236.1019(c).

The following limited operations exceptions apply to this MTEA:

 49CFR§236.1019(c)(1)(i) – the track is used for limited operations by at least one passenger railroad with all trains limited to restricted speed,

 49CFR§236.1019(c)(1)(iii) – not more than four passenger trains per day are operated on a segment of track of a Class 1 freight railroad on which less than 15 million gross tons of freight traffic is transported annually.

Each request in this MTEA is separately presented in detail in the following sections. Each has been reviewed and approved by Amtrak, which is the sole passenger train operator on these lines, and all are submitted jointly with Amtrak‟s full concurrence and agreement. Each request includes a summary track description and layout as well as a narrative description of the normal train operations and a reference to the applicable section of 49CFR§236.1019(c).

A main line track exception is requested by CN for each of the following track segments:

1. Freeport Sub / Chicago Sub / St. Charles Airline to reach Chicago Union Station

a) Freeport Sub from MP 2.9 to MP 2.1,

b) Chicago Sub from MP 2.2 to 1.4,

c) St. Charles Airline from 16th Street interlocking to connection with BNSF Railway at the end of CN‟s ownership approximately 70 feet west of the bascule bridge over the South Branch of the Chicago River (approximately 0.6 miles).

2. Memphis Sub MP 380.4 (Woodstock) to MP 394.3 (Y&MV Junction),

3. Y&MV Main from Y&MV Junction (Memphis Sub) to MP 5.4 (Shelby Sub),

4. Whirlpool Bridge – MP 0.0 to MP 0.6 Grimsby Subdivision

5. McComb Sub from MP 904.4 (Mays Yard) to MP 908.6 (Southport Junction).

6. Rouse‟s Point Sub from MP 1.18 (International Border) to MP 0.0 (Rouse‟s Point)

Detailed information for each request is provided in the sections below.

101

PTCIP

13.2. MTEA Request – Freeport/Chicago Subdivision and St. Charles Airline CN has two key routes that access downtown Chicago that are used by passenger train operations, these are the Freeport Subdivision and the Chicago Subdivision. The Freeport and Chicago subdivisions join at a station called 16th Street which is located at MP 2.1 of the Freeport Subdivision and MP 1.5 of the Chicago Subdivision. In addition, passenger trains in downtown Chicago move over CN‟s St. Charles Airline, which overlaps the Chicago Subdivision and extends beyond it to the West over the South Branch of the Chicago River.

CN is requesting a main line track exception for following three segments of track that join each other at the 16th Street interlocking:

a. Freeport Sub from MP 2.9 to MP 2.1 (16th Street interlocking) b. Chicago Sub from MP 2.2 to MP 1.4 (16th Street interlocking) c. St. Charles Airline from 16th Street interlocking to connection with BNSF Railway at the end of CN‟s ownership, which is approximately 70 feet west of the bascule bridge over the South Branch of the Chicago River (approximately 0.6 miles). 13.2.1. Freeport Subdivision There are no regularly scheduled passenger train operations on the section of the Freeport Subdivision between the 16th Street Interlocking to the north (MP 2.1), and Cermak Road (MP 2.9) to the south, which is the subject of this main line track exception request. When congestion arises from time to time on tracks entering/exiting Union Station, however, Amtrak and Metra trains may be re-routed across this section of track. In addition, there are only very limited freight operations on this segment of track, as CN uses this track just 3 days per week for local switching moves that service 2 customers.

Passenger operations on the Freeport Subdivision are primarily south of MP 2.9 (and thus are not part of this main line track exception request). There are 10 daily Amtrak trains that use CN‟s Freeport Subdivision south of MP 2.9 to access Union Station, including 2 trains (one daily round trip) for Amtrak‟s long haul Texas Eagle service and 8 short-haul trains (4 daily round trips) that are part of Amtrak‟s Lincoln Corridor service between Chicago and Carbondale/Quincy/St. Louis. These trains travel to and from Chicago Union Station on Amtrak-owned tracks, through the Lumber Street interlocking and across the South Branch bascule bridge, where they enter and exit CN‟s Freeport Subdivision at MP 2.9 (Cermak). South of MP 2.9 (Cermak) these trains travel on CN‟s Freeport Subdivision as far as MP 4.4 (Bridgeport) where they exit the Freeport Subdivision onto CN‟s Joliet subdivision. In addition to these Amtrak trains, there are 6 daily Metra trains that utilize CN‟s Freeport subdivision between MP 2.9 and MP 4.4.

Our main line track exception request includes the connecting track used by Amtrak that runs from 21st Street (MP 2.7) in the north to Cermak Road (MP 2.9). CN owns this connecting track, but the Amtrak dispatcher controls the signals governing access to the Amtrak tracks at 21st Street. All train operations on the connecting track between the Freeport Subdivision and Amtrak owned tracks are operated at restricted speed and there is a 10mph speed restriction on the Freeport main tracks between 16th and 21st Street stations (MP 2.1 to MP 2.7). Train operations are under the control of a TCS system between 16th and 21st Streets.

102

PTCIP

13.2.2. Chicago Subdivision There are 6 daily Amtrak trains that use CN‟s Chicago Subdivision, including one daily roundtrip for Amtrak‟s long-haul City of New Orleans service and 4 other short-haul trains that operate between Chicago and Carbondale Illinois.

Amtrak trains travel to and from the Chicago Union Station using the St. Charles Airline track that connects to CN‟s Chicago Subdivision at the 16th Street interlocking (MP 1.4 on the Chicago Subdivision). All train operations on the St. Charles Airline, and all train operations on the Chicago Subdivision between MP 1.4 (16th Street interlocking) and MP 2.2 have a maximum authorized speed of 10mph. Train operations on the Chicago subdivision are under the control of a TCS system from 16th Street station (MP 1.5) southward.

13.2.3. MTEA Request Based on the low volume of train operations and the existing low track speeds, for the three segments that are the subject of this request, CN is proposing to change the maximum train operating speed to Restricted Speed with a maximum speed of 10 mph and is requesting a main line track exception under 49CFR§236.1019(c)(1)(i) because the track is used for limited operations by at least one passenger railroad with all trains limited to restricted speed.

103

PTCIP

13.3. MTEA Request – Memphis Subdivision MP 380.4 to MP 394.3 Amtrak‟s City of New Orleans service, operating between Chicago and New Orleans, makes regularly scheduled stops in Memphis. To access the Amtrak Memphis Central Station, southbound Amtrak trains exit the CN Fulton subdivision at MP 380.4 (Woodstock) and use CN‟s Memphis subdivision from MP 380.4 (Woodstock) to MP 391.8 (Amtrak Memphis Central Station). Departing the Memphis Central Station trains operate south on the Memphis Subdivision from MP 391.8 to the junction with the Y&MV Main at MP 394.3 (Y&MV Jct.), where the southbound Amtrak trains exit CN‟s Memphis Subdivision. Similarly northbound Amtrak movements use the Memphis subdivision to go from MP 394.3 (Y&MV Jct) to the Memphis Central Station and then continue north on the Memphis Subdivision to Woodstock (MP80.4) where they access CN‟s Fulton Subdivision.

Annual freight traffic volume on CN‟s Fulton Subdivision north of Woodstock is 25 MGT, but there is very limited use of the northern part of the Memphis subdivision for freight operations. The northern portion of the Memphis Subdivision from Woodstock to Pyramid (MP 380.4 to MP 389.9) carries only 1.5 MGT of freight traffic. In fact, there are no freight operations at all on the Memphis subdivision between mileage posts 390 to 391.8 (Memphis Central Station).

The Memphis subdivision has CTC at the junction with the Fulton Sub and bi-directional ABS with track authorities required from MP 380.6 to MP 391.6. All train operations within the CTC and ABS portions of the Memphis subdivision are limited to 30 mph for passenger trains and 25 mph for freight trains. All train operations from MP 391.6 to the end of the Memphis Subdivision at MP 397.5 are restricted speed operations within yard limits with maximum speed restriction of 10 mph (MP 391.5 to MP 392.5) or 20 mph (MP 392.5 to MP 397.5).

Based on the low number of Amtrak movements and the limited freight operations on the northern portion of the Memphis Subdivision, a main line track exception is requested under 49CFR§236.1019(c)(3) because not more than four passenger trains per day are operated on a segment of track of a Class 1 freight railroad on which less than 15 million gross tons of freight traffic is transported annually.

See attached maps for clarification of the tracks impacted by this request.

104

PTCIP

105

PTCIP

13.4. MTEA Request – Y&MV Main Amtrak‟s City of New Orleans service, operating between Chicago and New Orleans, makes regularly scheduled stops in Memphis. To access the Memphis Central Station, northbound Amtrak trains use a short connecting track called the Y&MV Main to transfer from CN‟s Shelby Subdivision to the Memphis Subdivision. Amtrak trains exit the CN Shelby subdivision at MP 5.4 (West Jct.) and use the Y&MV Main track to connect to the Memphis Subdivision at MP 394.3 (Y&MV Jct.). Similarly southbound Amtrak movements use the Y&MV Main track when leaving Amtrak‟s Central Station to travel to CN‟s Shelby Subdivision at West Jct.

The Y&MV Main is not classified as main track and carries no freight train operations. This track has a maximum operating speed of 20mph and all operations are under the direction of the Memphis Yardmaster.

Based on the low number of Amtrak movements and the lack of any freight operations on the Y&MV Main, a main line track exception is requested under 49CFR§236.1019(c)(3) because not more than four passenger trains per day are operated on a segment of track of a Class 1 freight railroad on which less than 15 million gross tons of freight traffic is transported annually.

See the preceding map for clarification of the tracks impacted by this MTEA request

106

PTCIP

13.5. MTEA Request – Whirlpool Bridge Amtrak‟s Maple Leaf service, operating between New York City and Toronto, Canada uses the Whirlpool bridge to cross the Niagara River at Niagara Falls. Amtrak movements travel on CSX track to the southern side of the Whirlpool bridge where they connect to the CN Grimsby Subdivision at MP 0.0 and proceed northward onto the Whirlpool Bridge from MP 0.25 to MP 0.44.

There are two Amtrak trains per day that operate over the Whirlpool bridge to go from Niagara Falls, New York, to Niagara Falls, Ontario, Canada. The trains pass across the international boundary while on the bridge and travel to the VIA Railway train station at mile point 0.6 where they change crews to continue the trip to Toronto. There are no freight operations over the Whirlpool bridge, the track is classified as non-main track and all movements are made with a 10 mph speed restriction.

Train movements over the Whirlpool Bridge are controlled from CN‟s dispatch office located in Toronto, Ontario. Currently there is no plan to install PTC on the portion of the Grimsby Subdivision on the Canadian side of the bridge or on any other track dispatched from the Toronto dispatch office. Implementation of PTC over the bridge will require installation of PTC office computers in Toronto as well as development of a special interface to CN‟s Sieman‟s developed dispatch system.

Based on the low number of Amtrak movements and the lack of any freight operations on the Whirlpool Bridge, a main line track exception is requested under 49CFR§236.1019(c)(3) because not more than four passenger trains per day are operated on a segment of track of a Class 1 freight railroad on which less than 15 million gross tons of freight traffic is transported annually.

107

PTCIP

Figure 9 MTEA Request – Whirlpool Bridge

108

PTCIP

13.6. MTEA Request – McComb Subdivision MP 904.4 to MP 908.6 Mays Yard is the southernmost destination for the vast majority of freight traffic on CN‟s McComb subdivision that extends from Jackson Mississippi to New Orleans. The McComb Subdivision north of Mays Yard has an annual freight traffic volume of 10.7 MGT while the track between Mays Yard and Southport Jct (4.2 miles) to the south carries only 0.3 MGT of freight traffic annually. Train operations are under TCS control from MP 904.4 to MP 906.1 and Manual Interlocking Limits from MP 906.4 to MP 908.6.

There are four daily Amtrak trains that operate on the CN McComb subdivision as follows:

1. The daily Amtrak train “City of New Orleans” that operates between Chicago and New Orleans uses the track south of Mays Yard to connect to the NOUPT (New Orleans Union Passenger Terminal) track that provides access to the New Orleans Amtrak station,

2. The daily Amtrak train “Sunset Limited” that operates between Los Angeles and New Orleans connects to CN track at East Bridge Jct (MP 906.4) and travels south to connect to the NOUPT (New Orleans Union Passenger Terminal) track that provides access to the New Orleans Amtrak station.

In summary, 2 daily Amtrak trains between MP 904.4 and MP 906.6 and 4 daily Amtrak trains travel between MP 906.6 and MP 908.6 of the McComb subdivision. Based on the low volume of freight operations and the limited number of Amtrak passenger trains, a main line track exception is requested under 49CFR§236.1019(c )(3) because not more than four passenger trains per day are operated on a segment of track of a Class 1 freight railroad on which less than 15 million gross tons of freight traffic is transported annually.

The attached map below provides a view of the track covered by this MTEA request.

Figure 10 MTEA Request – McComb Sub MP904.4 to MP908.6

13.7. MTEA Request – Rouse’s Point Sub MP 1.18 to 0.0 Amtrak‟s Adirondack service, operating between New York City and Montreal, Canada, uses the CN Rouse‟s Point Subdivision to travel from Rouse‟s Point, New York to Montreal, Canada. The CN Rouse‟s Point subdivision is 42.7 miles long and joins with the CN St. Hyacinthe subdivision for the final access to Montreal. There are 1.18 miles on CN‟s Rouse‟s Point subdivision that is within the United States and subject to PTC regulation. This portion runs from downtown Rouse‟s Point, NY, to the end of the Rouse‟s Point Subdivision (MP 0.0), at the international border.

109

PTCIP

There are two Amtrak trains per day that operate over the Rouse‟s Point subdivision between Montreal and downtown Rouse‟s Point. There are no freight operations on the US portion of the Rouse‟s Point subdivision, the track is classified as non-main track, and all movements are made with a 10 mph speed restriction.

Train movements on the Rouse's Point Subdivision are controlled from CN‟s dispatch office located in Montreal, Quebec. Currently there is no plan to install PTC on the portion of the Rouse's Point Subdivision on the Canadian side of the border or on any other track dispatched from the Montreal dispatch office. Implementation of PTC for the US portion of the Rouse's Point Subdivision will require installation of PTC office computers in Montreal as well as development of a special interface to CN‟s Sieman‟s developed dispatch system.

Based on the low number of Amtrak movements and the lack of any freight operations on the US portion of the Rouse‟s Point subdivision, a main line track exception is requested under 49CFR§236.1019(c)(3) because not more than four passenger trains per day are operated on a segment of track of a Class 1 freight railroad on which less than 15 million gross tons of freight traffic is transported annually.

110

PTCIP

Figure 11 MTEA Request – Rouse’s Point Sub MP 1.18 to 0.0

111

PTCIP

14. De Minimis Track Exclusion Requests [§236.1005]

14.1. De Minimis General Section §236.1005(b)(4)(ii) of the final PTC rule provides an exception to PTC implementation for lines with de minimis TIH/PIH risk. The section allows railroad to request review of the requirement for installation of PTC on a low density track segment where PTC would otherwise be required but has not yet been installed.

The rule establishes criteria that must be satisfied for a de minimis track exclusion request to be considered. The primary de minimis criteria for track segments to be considered for an exclusion request is that it carry less than 100 cars of TIH/PIH material (load and residue) per year §236.1005(b)(4)(ii)(A).

If a track segment meets the minimum TIH/PIH carload criteria and absent of any other special circumstance, an exclusion request will typically be granted if the following additional criteria are met §236.1005(b)(4)(ii)(B).

1. Track consists of Class1 or Class 2 track - §236.1005(b)(4)(ii)(B)(1), 2. Track carries less than 15 MGT annually - §236.1005(b)(4)(ii)(B)(2), 3. Track has a ruling grade less than 1 percent - §236.1005(b)(4)(ii)(B)(3), 4. Temporal separation of TIH/PIH traffic - §236.1005(b)(4)(ii)(B)(4).

For line segments that meet the minimum 100 TIH/PIH carload criteria as well as the maximum 15 MGT total traffic volume criteria but do not meet the other de minimis exclusion criteria (track class, grade and temporal separation) an exclusion request may still be submit for consideration if the railroad can show that risk mitigations will be applied that will ensure the risk of release of TIH/PIH materials is negligible §236.1005(b)(4)(ii)(C).

The issue of negligible risk of TIH/PIH release is one that CN wishes to discuss further with the FRA. CN believes that there is a need to develop and specify a common methodology and tools that can be used by the railway and the regulatory agencies to assess and define what constitutes negligible risk. CN is willing to commit time and resources to working with the FRA and other parties as appropriate to achieve this objective. It is CN‟s intention to use the established tools and methodology to review and confirm the de minimis exclusion requests being put forward in this implementation plan.

CN has several track segments that are less than 15 MGT of annual freight traffic with minimal volumes of TIH/PIH and no passenger operations that meet the criteria for a de minimis based review of PTC implementation requirement. Each track segment proposed for a de minimis based PTC requirement review is presented separately in the following sections. Each request includes a map of the track covered, a summary track description and layout as well as a narrative description of the normal train operations.

112

PTCIP

The proposed track segments are:

1. Cherokee Subdivision – MP 381.3 to MP 508.8 – 105.1 track miles, 2008 traffic was 5.7 MGT with 29 carloads of TIH/PIH traffic. 2. Minneapolis Subdivision – MP 308.5 to MP 432.1 – 123.6 track miles, 2008 traffic was 5.7 MGT with 24 carloads of TIH/PIH traffic,

The Minneapolis Subdivision being proposed for de minimis based review of PTC implementation requirement ha sdropped below the 5 MGT threshold for main line track segments in 2009. If traffic volumes in 2010 remain below 5 MGT, the Minneapolis subdivision line segment will be a candidate for review of PTC requirements exclusion under §236.1005(b)(4)(i).

113

PTCIP

14.2. De Minimis Request – Cherokee Subdivision The CN Cherokee Subdivision is 105.1 miles long and extends westward from Tara (MP 381.2) to its termination at Sioux City Iowa (MP 508.8). This subdivision is the western end of the CN line that runs from Chicago to Sioux City Iowa and at it‟s eastern end joins with the CN Omaha Subdivision that extends southwest to the city of Council Bluffs Iowa. Traffic volumes on the Cherokee subdivision are significantly lower than its neighbouring subdivision to the east (Waterloo Subdivision) due the division of traffic between the Omaha and Cherokee subdivisions.

Train operations on the Cherokee Subdivision typically consist of 2 trains per day (1 eastbound and 1 westbound move) that operate between Chicago and Sioux City Iowa, with no passenger train operations. Following are some of the key operational and traffic metrics for the Cherokee Subdivision:

1. Track Miles: MP 381.2 to MP 508.8 – 127.6 track miles, 2. Traffic Volume: 2008 - 5.7 MGT, 2009 – 5.2 MGT, 3. Train Starts: 2008 – 471, 2009 – 434 4. Train Miles: 2008 – 122,924 miles , 2009 – 109,963 miles 5. TIH Traffic Volume: 2008 - 29 cars (18L, 11R), 2009 – 6 cars (0L, 6R), 6. Track Class: Class 3 maximum speed 40 mph, 7. Ruling Grade: 1.2% ascending westbound grade,, 8. Method of Operations: 105.1 miles of TWC, 9. Curvature: 54 curves, max. curvature 3.3 degrees, avg. 0.5 curves/mile, 10. Crossings: 256 road crossings, average of 2.43 per mile 11. Turnouts: 54 mainline turnouts, average of 0.5 per mile. 12. Summary weighted risk ranking: 1.950 tied for 36th out of 41 CN subdivisions that are required to install PTC.

CN operates a straight line railway network, with a single path between Chicago and Sioux City Iowa and has no plans for TIH/PIH or other traffic re-routing onto or away from the Cherokee Subdivision for the next 5 years. In addition CN is not aware of any new or pending customer agreements or traffic changes on this corridor that would impact the volume of TIH/PIH shipments over the next 5 years.

Traffic volume data for the Cherokee sub is available for 9 discrete measurement sections on the Cherokee Sub and the weighted average traffic volume is 5.7MGT in 2008 and 5.2MGT in 2009. The highest traffic volumes are on the east end of the subdivision with peak volumes of 6.3MGT in 2008 and 6.0MGT in 2009. Lightest traffic volumes are on the west end of the subdivision with traffic volumes of 5.0MGT in 2008 and 4.5MGT in 2009.

114

PTCIP

Following is a comparison of the Cherokee Subdivision traffic and track attributes with the established criteria for a de minimis review of PTC implementation:

1. TIH Volume: Cherokee 29 TIH/PIH cars in 2008 – meets criteria, 2. Track Class: Cherokee maximum speed 40 mph – exceeds criteria by 15 mph, 3. Traffic Volume: Cherokee volume 5.7MGT in 2008 – meets criteria, 4. Ruling Grade: Cherokee maximum grade 1.2% - exceeds criteria by 0.2% 5. Temporal Separation: Cherokee sub had 2 trains per day but temporal separation is not guaranteed.

The Cherokee subdivision exceeds the minimum requirements for de minimis PTC exclusion in 3 areas (train speed, ruling grade and temporal separation) and based on this CN is providing the following proposed risk mitigations to ensure there is a negligible risk of TIH/PIH release if PTC is not installed on the Cherokee subdivision.

Risk Mitigations 1. Reduce speed of trains carrying TIH/PIH traffic to 25mph on the Cherokee Subdivision to ensure TIH/PIH train speeds are equal to the maximum allowed speeds for Class 2 track, 2. Reduce speed of trains not carrying TIH/PIH to 25mph when passing trains carrying TIH/PIH on the Cherokee Subdivision, 3. All CN train movements on the Cherokee Subdivision are already operated in full compliance with the requirements of AAR Circular No. OT-55 (Recommended Railroad Operating Practices for Transportation of Hazardous Materials). CN will maintain its compliance with these voluntary recommended operating practices.

Speed Restriction Enforcement To ensure compliance with the speed restrictions identified as risk mitigation measures, CN proposes to use the following strategies: 1. Issue a 25 mph maximum speed mandatory directive to the TIH/PIH carrying train for the entire Cherokee Subdivision,

2. Issue instructions to Rail Traffic Controllers that TIH/PIH carrying trains on the Cherokee Subdivision are to hold the main track at all passing sidings unless it is operationally impractical to do so. This will ensure that opposing or passing trains using the siding track must reduce their speeds while passing the TIH/PIH carrying train,

In addition to the above CN will work with the PTC System equipment developers to evaluate the feasibility and implement, if possible, additional functionality to the PTC system that will enable positive enforcement of the 25mph mandatory speed directives using the capabilities of the on board PTC computer systems.

CN is confident that these proposed mitigation measures will reduce that the probability of a TIH/PIH product release to the required “negligible risk” level required to be considered for

115

PTCIP relief of PTC implementation. To ensure that that the “negligible risk” level is achieved, CN commits to performing a risk review of the PTC required track on the Cherokee Subdivision when industry standard tools and methodology are available. Furthermore, CN commits to performing an annual review of Cherokee Subdivision traffic volumes and operations as part of the annual PTC update process to FRA with specific emphasis on TIH/PIH traffic. The annual review shall include re-affirmation of that there remains a “negligible risk” of TIH/PIH release

A map of the Cherokee subdivision with the de minimis track highlighted is attached below.

Figure 12 De Minimis Request – Cherokee Subdivision

116

PTCIP

14.3. De Minimis Request – Minneapolis Subdivision The CN Minneapolis Subdivision is 123.6 miles long and extends westward from the junction with CN‟s Superior subdivision at Owen (MP 308.5) to its termination at Withrow (MP 432.5). At the west end of the subdivision most traffic connects to CP or BN for connection to customers in Minneapolis/St. Paul. A small percentage of the traffic connects to CN‟s Dresser subdivision. Train operations on the Minneapolis Subdivision typically consist of 2 to 3 trains per day with no passenger operations. Following are some of the key operational and traffic metrics for the Cherokee Subdivision:

1. Track Miles: MP 308.5 to MP 432.1 – 123.6 track miles, 2. Traffic Volume: 2008 - 5.7 MGT, 2009 – 3.8 MGT, 3. Train Starts: 2008 – 500, 2009 – 193 4. Train Miles : 2008 – 146,303 miles , 2009 – 113,995 miles 5. TIH Traffic Volume: 2008 - 24 cars (10L, 14R), 2009 – 1 car (1L, 0R), 6. Track Class: Class 3 maximum speed 40 mph, 7. Ruling Grade: 1.1% ascending eastbound grade, 8. Method of Operations: 123.6 miles of TWC, 9. Curvature: 43 curves, max. curvature 3.82 degrees, avg. 0.35 curves/mile, 10. Crossings: 187 road crossings, average of 1.51 per mile 11. Turnouts: 41 mainline turnouts, average of 0.3 per mile. 12. Summary weighted risk ranking: 1.950 tied for 36th out of 41 CN subdivisions that are required to install PTC.

CN operates a straight line railway network, with a single path from our Superior Subdivision at Owen towards Minneapolis. There are no plans for TIH/PIH or other traffic re-routing onto or away from the Minneapolis Subdivision for the next 5 years. In addition CN is not aware of any new or pending customer agreements or traffic changes on this corridor that would impact the volume of TIH/PIH shipments over the next 5 years. Following is a comparison of the Minneapolis Subdivision traffic and track attributes with the established criteria for a de minimis review of PTC implementation:

1. TIH Volume: Minneapolis 24 TIH/PIH cars in 2008 – meets criteria, 2. Track Class: Minneapolis max, speed 40 mph – exceeds criteria by 15 mph, 3. Traffic Volume: Minneapolis volume 5.7MGT in 2008 – meets criteria, 4. Ruling Grade: Minneapolis max. grade 1.1% - exceeds criteria by 0.1% 5. Temporal Separation: Minneapolis sub has limited train operations per day but temporal separation is not guaranteed.

117

PTCIP

The Minneapolis subdivision exceeds the minimum requirements for de minimis PTC exclusion in 3 areas (train speed, ruling grade and temporal separation) and based on this CN is providing the following proposed risk mitigations to ensure there is a negligible risk of TIH/PIH release if PTC is not installed on the Minneapolis subdivision.

Risk Mitigations 1. Reduce speed of trains carrying TIH/PIH traffic to 25mph on the Minneapolis Subdivision to ensure TIH/PIH train speeds are equal to the maximum allowed speeds for Class 2 track, 2. Reduce speed of trains not carrying TIH/PIH to 25mph when passing trains carrying TIH/PIH on the Minneapolis Subdivision, 3. All CN train movements on the Minneapolis Subdivision are already operated in full compliance with the requirements of AAR Circular No. OT-55 (Recommended Railroad Operating Practices for Transportation of Hazardous Materials). CN will maintain its compliance with these voluntary recommended operating practices.

2. Issue instructions to Rail Traffic Controllers that TIH/PIH carrying trains on the Minneapolis Subdivision are to hold the main track at all passing sidings unless it is operationally impractical to do so. This will ensure that opposing or passing trains using the siding track must reduce their speeds while passing the TIH/PIH carrying train. In addition to the above CN will work with the PTC System equipment developers to evaluate the feasibility and implement, if possible, additional functionality to the PTC system that will enable positive enforcement of the 25mph mandatory speed directives using the capabilities of the on board PTC computer systems.

CN is confident that these proposed mitigation measures will reduce that the probability of a TIH/PIH product release to the required “negligible risk” level required to be considered for relief of PTC implementation. To ensure that that the “negligible risk” level is achieved, CN commits to performing a risk review of the PTC required track on the Minneapolis Subdivision when industry standard tools and methodology are available. Furthermore CN commits to performing an annual review of Minneapolis Subdivision traffic volumes and operations as part of the annual PTC update process to FRA with specific emphasis on TIH/PIH traffic. The annual review shall include re-affirmation of that there remains a “negligible risk” of TIH/PIH release.

A map of the Minneapolis subdivision with the de minimis track highlighted is attached below.

118

PTCIP

Figure De Minimis Request – Minneapolis Subdivision

Appendix A: Line Segment Attributes Detailed Tables

119

PTCIP

Appendix A: Line Segment Attributes Detailed Tables

TABLE 45 TRAFFIC CHARACTERISTICS BY DEPLOYMENT GROUP ...... 121 TABLE 46 OPERATING CHARACTERISTICS BY DEPLOYMENT GROUP ...... 122 TABLE 47 TRACK ATTRIBUTES TABLE ...... 123 TABLE 48 DEPLOYMENT GROUP ATTRIBUTES – RAILWAY CROSSINGS ...... 123 TABLE 49 PASSENGER TRAIN OPERATIONS ...... 126 TABLE 50 METRA PASSENGER TRAIN SUMMARY ...... 126 TABLE 51 AMTRAK PASSENGER TRAIN SUMMARY ...... 127

120

PTCIP

Table 45 Traffic Characteristics by Deployment Group

Deployment Subdivision 2008 Traffic Data Group Avg TIH/PIH Other Psgr MGT Peak MGT Cars Hazmat Trains Pilot Baton Rouge 11.5 15.0 22,815 112,608 0 McComb 24.2 31.2 17,852 128,252 2 Central Cairo 22.4 22.5 4,139 33,595 2 Centralia 29.4 39.5 4,288 34,816 6 Champaign 39.6 51.5 9,344 106,644 6 Chicago 39.7 43.8 8,866 156,239 6 Fulton 40.1 50.2 10,846 83,910 2 Gilman 2.4 5.5 589 387 0 Peoria 5.1 8.0 2,247 6,095 0 St. Louis 8.3 12.9 29 2,233 0 Gulf Beaumont 11.2 15.6 8,565 27,146 0 Bluford 26.1 46.6 11,035 114,279 0 Hammond 5.4 5.4 4,524 15,534 0 Memphis 15.2 49.8 11,305 108,284 2 P&I Railroad 46.6 46.6 2,689 14,658 0 Shelby 20.9 24.0 10,846 83,910 2 Yazoo 45.2 48.4 9,099 98,003 2 East Cherokee 5.7 6.3 29 12,430 0 Dubuque 11.6 14.1 649 39,132 0 Elsdon 26.0 26.0 6,655 136,562 2 Flat Rock 7.8 13.2 1,447 17,318 0 Flint 42.4 53.4 8,509 151,187 2 Freeport 9.9 11.8 1,997 92,319 16 Holly 7.0 9.0 3,399 46,716 6 Joliet 4.3 6.8 4,215 97,290 16 Mt. Clemens 17.3 17.3 6,753 135,908 0 Shoreline 16.2 41.9 2,964 43,412 6 South Bend 42.3 51.9 6,016 108,708 8 Waterloo 9.7 9.7 463 31,610 0 North Fox River 9.1 20.0 27 11,185 0 Leithton 19.3 43.5 2,429 72,623 0 Matteson 11.5 44.4 5,971 108,411 0 Minneapolis 5.7 5.7 24 6,793 0 Missabe 36.3 74.5 729 35,988 0 Neenah 52.9 57.1 2,590 81,033 0 Rainy 38.5 44.5 1,451 84,376 0 Sprague (US) 44.7 46.0 1,531 86,352 0 Superior 39.8 55.2 2,277 96,883 0 Valley 3.2 5.5 863 8,735 0 Waukesha 52.3 61.6 2,508 74,567 22

121

PTCIP

Table 46 Operating Characteristics by Deployment Group

Deployment Subdivision Method of Operations Miles of Main Track Max Group Train ABS/ Yard/ Total Total Speed TCS TWC TWC 520 Miles Main1 Main2 Main3 Main4 Miles Pilot Baton Rouge 0.7 76.8 1.9 79.4 79.4 79.4 40.0 McComb 209.2 3.7 5.4 218.3 181.4 36.9 218.3 79.0 Central Cairo 39.5 4.4 43.9 41.7 2.2 43.9 79.0 Centralia 142.7 142.7 115.9 26.8 142.7 79.0 Champaign 131.2 131.2 123.1 8.1 131.2 79.0 Chicago 171.2 171.2 121.9 35.6 8.1 5.6 171.2 79.0 Fulton 133.0 3.2 136.2 118.7 17.5 136.2 79.0 Gilman 28.9 28.9 28.9 28.9 60.0 Peoria 2.0 38.1 40.1 40.1 40.1 40.0 St. Louis 37.1 37.1 37.1 37.1 60.0 Gulf Beaumont 168.8 168.8 168.8 168.8 49.0 Bluford 121.7 41.9 163.6 163.6 163.6 60.0 Hammond 42.7 42.7 42.7 42.7 49.0 Memphis 2.5 2.5 2.5 2.5 30.0 P&I Railroad 13.0 13.0 13.0 13.0 40.0 Shelby 6.7 16.1 10.4 33.2 16.6 16.6 33.2 60.0 Yazoo 202.7 2.8 205.5 204.1 1.4 205.5 79.0 East Cherokee 127.6 127.6 127.6 127.6 40.0 Dubuque 133.7 133.7 133.7 133.7 50.0 Elsdon 33.2 0.4 33.6 16.8 16.8 33.6 60.0 Flat Rock 23.9 23.9 23.9 23.9 55.0 Flint 227.4 227.4 155.6 68.7 3.1 227.4 79.0 Freeport 20.4 103.6 124.0 112.2 11.8 124.0 50.0 Holly 75.1 14.0 89.1 62.9 26.2 89.1 70.0 Joliet 66.4 66.4 33.2 33.2 66.4 79.0 Mt. Clemens 2.1 43.5 45.6 45.6 45.6 49.0 Shoreline 46.9 3.6 3.4 53.9 51.0 2.9 53.9 40.0 South Bend 260.2 260.2 142.5 117.7 260.2 60.0 Waterloo 103.8 103.8 103.8 103.8 50.0 North Fox River 6.3 28.6 34.9 34.9 34.9 49.0 Leithton 70.4 6.4 76.8 65.0 11.8 76.8 45.0 Matteson 18.4 48.7 67.1 43.4 23.7 67.1 45.0 Minneapolis 123.6 123.6 123.6 123.6 40.0 Missabe 36.2 34.3 70.5 63.5 7.0 70.5 49.0 Neenah 95.2 95.2 86.6 8.6 95.2 60.0 Rainy 92.1 60.1 152.2 152.2 152.2 60.0 Sprague (US) 43.4 43.4 43.4 43.4 60.0 Superior 233.3 233.3 233.3 233.3 60.0 Valley 1.0 11.3 12.3 12.3 12.3 40.0 Waukesha 184.9 184.9 141.7 37.7 5.5 184.9 60.0

122

PTCIP

Table 47 Track Attributes Table Group Subdivision Track Attribute Road Crossings Switches Max Curves Route Track Grade Miles Miles Total Controlled Hand Num/Mile Total Num/Mile (Percent) Total Max(Deg) Switches Throw Switches

Pilot Baton Rouge 250 1 97 3.1 98 1.2 0.50 51 7.63 79.4 79.4 McComb 293 28 85 1.6 113 0.5 0.45 165 3.91 181.4 218.3 Central Cairo 61 7 10 1.5 17 0.4 0.74 42 4.74 41.7 43.9 Centralia 161 19 51 1.4 70 0.5 0.84 113 5.07 115.9 142.7 Champaign 167 18 57 1.4 75 0.6 0.50 52 2.49 123.1 131.2 Chicago 169 67 66 1.4 133 0.8 0.55 55 4.11 121.9 171.2 Fulton 167 20 53 1.4 73 0.5 0.51 164 4.11 118.7 136.2 Gilman 58 6 13 2.0 19 0.7 1.00 9 2.32 28.9 28.9 Peoria 76 0 14 1.9 14 0.3 1.12 30 6.18 40.1 40.1 St. Louis 61 7 14 1.6 21 0.6 0.95 20 5.40 37.1 37.1 Gulf Beaumont 301 0 87 1.8 87 0.5 1.00 224 9.95 168.8 168.8 Bluford 98 18 20 0.6 38 0.2 0.30 78 2.56 163.6 163.6 Hammond 152 0 35 3.6 35 0.8 0.50 29 11.49 42.7 42.7 Memphis 4 0 6 1.6 6 2.4 2.00 6 4.04 2.5 2.5 P&I Railroad 14 4 10 1.1 14 1.1 2.00 6 1.96 13.0 13.0 Shelby 25 0 30 1.5 30 0.9 0.49 21 2.18 16.6 33.2 Yazoo 330 32 53 1.6 85 0.4 0.50 179 5.05 204.1 205.5 East Cherokee 256 1 53 2.0 54 0.4 1.17 54 3.30 127.6 127.6 Dubuque 220 24 32 1.6 56 0.4 1.40 155 11.93 133.7 133.7 Elsdon 32 23 4 1.9 27 0.8 0.43 23 3.14 16.8 33.6 Flat Rock 69 5 6 2.9 11 0.5 0.30 27 6.82 23.9 23.9 Flint 277 48 50 1.8 98 0.4 0.60 75 4.00 155.6 227.4 Freeport 99 26 72 0.9 98 0.8 0.95 75 4.86 112.2 124.0 Holly 116 33 23 1.8 56 0.6 0.65 45 7.33 62.9 89.1 Joliet 45 23 38 1.4 61 0.9 0.34 12 4.04 33.2 66.4 Mt. Clemens 121 24 39 2.7 63 1.4 0.59 9 1.67 45.6 45.6 Shoreline 76 19 8 1.5 27 0.5 0.75 49 4.67 51.0 53.9 South Bend 285 43 36 2.0 79 0.3 0.88 73 7.58 142.5 260.2 Waterloo 235 28 57 2.3 85 0.8 1.00 70 6.46 103.8 103.8 North Fox River 105 4 40 3.0 44 1.3 0.65 33 5.72 34.9 34.9 Leithton 86 29 38 1.3 67 0.9 0.70 10 10.00 65.0 76.8 Matteson 66 25 53 1.5 78 1.2 0.87 53 8.00 43.4 67.1 Minneapolis 187 0 41 1.5 41 0.3 1.08 43 3.82 123.6 123.6 Missabe 22 19 18 0.3 37 0.5 1.08 67 8.30 63.5 70.5 Neenah 145 18 42 1.7 60 0.6 1.03 85 5.89 86.6 95.2 Rainy 91 9 28 0.6 37 0.2 1.20 123 5.00 152.2 152.2 Sprague (US) 66 12 4 1.5 16 0.4 0.50 8 2.19 43.4 43.4 Superior 279 53 59 1.2 112 0.5 1.90 152 5.51 233.3 233.3 Valley 29 0 15 2.4 15 1.2 1.04 10 13.95 12.3 12.3 Waukesha 250 91 90 1.8 181 1.0 1.08 196 7.23 141.7 184.9

Note 1) Within the Eldorado Subdivision, PTC implementation will not occur between Eldorado Junction and Akin Junction. See Section 8.3.3.1. Table 48 Deployment Group Attributes – Railway Crossings Deployment Railway Crossings at Grade

123

PTCIP

Group Mntce Subdivision Mileage CN Speed Other RR Description RR Pilot Baton Rouge 361.2 20 KCS KCS MP 361.2 (aka Scotland, automatic) McComb 908.6 60/40 KCS CN Southport Jct Central Centralia 253.2 35/35 BNSF, NS BNSF Centralia 266.3 50/30 CSX CN Ashley (automatic) Champaign 127.8 30/30 NS NS Champaign 137.1 50/30 NS CN Tolono 149.8 50/30 CSX, UP BNSF Tuscola 199.3 40/30 CSX CSX Effingham 229.4 60/40 UP BNSF Kinmundy 244.3 60/40 CSX CN Odin (automatic) Chicago 1.5 10-Oct Metra Metra 16th Street 14.5 25/25 CSSSB CSSSB Kensington 55.2 50/30 NS CN Kankakee 81.1 60/40 TP&W CN Gilman Gilman 81.1 10 TP&W CN Gilman 110 25 NS NS Gibson City 148.5 25 CN CN Clinton 169.2 25 CN CN Mt. Pulaski 190.3 25 NS CN Starnes Peoria 44.2 20 UP UP Lincoln 55.3 10 CN CN Mt. Pulaski 77.1 20 NS CN Wabic 103.2 40 UP UP Sullivan (automatic) St. Louis 61.1 60 UP CN Pinckneyville (automatic) Gulf Beaumont 0.5 10 NS NS Mobile (Rule 513 in Rule 520) 95 10 NS CN Hattiesburg (key release in YL) Bluford 2.2 30 PAL CN PAL Crossing (automatic) Baton Rouge Jct. (gates in Rule Hammond 0.8 10 KCS CN 520) Memphis 389.9 30/25 MATA CN Pyramid BNSF, 392.5 10-Oct CN Jct CSX, UP East Elsdon 11.8 15 BRC BRC Hayford 12.8 40 Metra Metra Ashburn 19.3 40 IHB IHB BI Jct 25.2 30/30 UP CN Thornton Jct 34 30/30 NS NS Hays 36.1 40/40 CN CN Griffith Flat Rock 6.4 20 CR CN MA-2 (automatic) 11.1 30 CN, CR CN FN 24.1 30 CSX CSX Carleton 37.7 25 AA CN Diann Flint 221.4 45/40 NS CN Cedar 223.5 50/40 CSX CN Trowbridge 253.2 45/45 CN CN Durand 271.8 55/30 CSX CSX Kearsley Freeport 2.1 10-Oct Metra CN 16th Street

124

PTCIP

Railway Crossings at Grade Deployment Mntce Group Subdivision Mileage CN Speed Other RR Description RR 2.7 10-Oct Amtrak CN 21st Street 5.6 15 CSX CN Ash Street 7.1 50 BNSF CN IN Crossing (automatic) 8.3 50 BRC CN Belt Crossing Freeport 85.6 25 DM&E, IR CN Rockford Holly 4.2 10-Oct CR CN Milwaukee Jct 46.3 40/40 CSX CN Holly 67 25/25 CN CN Durand CSX, Joliet 5.1 30/25 CSX, NS CP Brighton NS 6.6 30/30 BNSF BNSF Corwith 7.9 50/30 BRC BRC LeMoyne 13.1 50/30 CSX, IHB CSX CP Canal (aka ARGO) Mt. Clemens 7 25 CR CR Forest Lawn (automatic) Shoreline 1 10 CSX CSX Boulevard (Rule 520) 18.7 40/40 NS CN Warner 37.3 40/20 CN, CR CN FN 46.8 40/20 CR CR Victoria Ave 54 35/25 CR CR Beaubien St South Bend 52.6 45 CSX, NS CN Wayne 71.1 40 CSX CSX Wellsboro 80.2 40 CSSSB CN Stillwell 146.8 30 GDLK CN Schoolcraft Waterloo 315.7 30 CN CN Ackley 325.6 25 UP UP Mills 355.5 25 UP UP Webster City 381.2 25 UP UP Tara (gate) North Leithton 28.9 30 UP UP JB Tower 37.5 45 CP CN Spaulding 49.6 40 UP CN Barrington 60.3 45 CN CN Leithton 65.5 25 CP CN Rondout 67.1 25 UP CN Upton (automatic) Matteson 0.8 40 CSX, Metra CN E Joliet (aka Rock Island) 25.2 45 UP CN Chicago Heights 31.3 20 CSX CSX Dyer (automatic) 33.7 45 NS CN Hartsdale (automatic) 36.2 25 CN CN Griffith 39.8 45 NS CN Van Loon 41.8 45 IHB, NS CSX Ivanhoe Minneapolis 350.1 25 UP CN CF Yard (automatic) Missabe 68.4 25 CN CN Shelton Jct (automatic) Rainy 66.8 25 CN CN Ramshaw (automatic) 70.8 25 CN CN Shelton Jct (automatic) Superior 261 40 CN CN JO Valley 63.3 20 CN CN JO

125

PTCIP

Railway Crossings at Grade Deployment Mntce Group Subdivision Mileage CN Speed Other RR Description RR Waukesha 15.5 25/25 CP, Metra CP Tower B-12 23.3 25/25 UP CN Deval 37.9 60/50 CN CN Leithton 44 60/50 CP CP GraysLake 102.5 45 CP CP S Duplainville

Table 49 Passenger Train Operations Passenger Train Summary - Annual

Subdivision Amtrak Metra Total Passenger Total Total Miles Total Total Miles Miles (Est.) Trains (Est.) Trains (Est.)

Cairo 730 30,879 30,879 Centralia 2,920 170,455 170,455 Champaign 2,190 269,589 269,589 Chicago 2,190 268,494 268,494 Elsdon 730 4,307 4,307 Flint 730 113,588 113,588 Freeport 3,650 6,205 2,190 2,659 8,864 Fulton 730 81,176 81,176 Grimsby 730 876 876 Holly 2,190 47,523 47,523 Joliet 3,650 123,370 2,190 52,873 176,243 McComb 730 132,422 132,422 Memphis 730 10,147 10,147 Rouse's Point 730 438 438 Shelby 730 5,621 5,621 Shoreline 2,190 7,884 7,884 SouthBend 3,650 4,891 4,891 Waukesha 0 7,848 225,335 225,335 Y&MV Main 730 3,942 3,942 Yazoo 730 150,015 150,015 Total Train Miles 1,431,822 280,868 1,712,690

Table 50 Metra Passenger Train Summary

Metra Passenger Train Summary Subdivision From Station From To Station To Metra Daily Miles Metra MP MP Trains Miles/Yr (one-way) (Est.) Freeport Chicago (21st St.) 2.7 Bridgeport 4.4 6.0 1.7 2,659 Joliet Bridgeport 3.5 Joliet (UD Tower) 37.3 6.0 33.8 52,873 Waukesha Franklin Park (B12) 15.5 Antioch 55.7 21.5 40.2 225,335 Totals 75.7 280,868

126

PTCIP

Table 51 Amtrak Passenger Train Summary

Amtrak Passenger Train Summary Subdivision From Station From To Station To Amtrak Miles Amtrak MP MP Daily Trains Miles/Yr (one-way) (Est.) Cairo Illinois 363.1 Cairo Jct 405.4 2 42.3 30,879 Centralia Sandoval Jct 247.2 Carbondale 306 6 58.8 128,772 Centralia Carbondale 306.0 Illinois 363.1 2 57.1 41,683 Champaign Leverett Jct 124.1 Sandoval Jct 247.2 6 123.1 269,589 Chicago Clark St (16th St) 1.5 Leverett Jct. 124.1 6 122.6 268,494 Elsdon UP Thornton Jct. 25.2 CSX Maynard (Munster) 31.0 2 5.9 4,307 Flint Emmett St. 178.6 Port Huron 334.2 2 155.6 113,588 Freeport Chicago (21st St.) 2.7 Bridgeport 4.4 10 1.7 6,205 Fulton Cairo Jct 269.2 Woodstock 380.4 2 111.2 81,176 Grimsby Rouse's Point 0.0 International Border 1.2 2 1.2 876 Holly Milwaukee Jct 4.1 Pontiac 25.8 6 21.7 47,523 Joliet Bridgeport 3.5 Joliet (UD Tower) 37.3 10 33.8 123,370 McComb Jackson 727.2 Southport Jct. 908.6 2 181.4 132,422 Memphis Woodstock 380.4 West Jct 394.3 2 13.9 10,147 Rouse's Point Bridge 0.0 International Border 0.6 2 0.6 438 Shelby West Jct 5.4 Lakeview 13.1 2 7.7 5,621 Shoreline CP-Vinewood St 51.2 Milwaukee Jct 54.8 6 3.6 7,884 SouthBend Gord 175.5 Baron 176.7 8 1.2 3,504 SouthBend Baron 176.7 Emmett St 178.6 2 1.9 1,387 Y&MV Main Y&MV Jct 0.0 West Jct 5.4 2 5.4 3,942 Yazoo Lakeview 13.1 Jackson 218.6 2 205.5 150,015 Totals 1,156.2 1,431,822

127

PTCIP

APPENDIX B: RISK FACTOR PRIORITIZATION MODEL

128

PTCIP

Appendix B: Risk Factor Prioritization Model

TABLE OF CONTENTS Page 1.0 Introduction ...... 131 2.0 Risk Prioritization Model Approach ...... 132 2.1 Identification of Risk Factors ...... 132 2.2 Estimation of Risk Factor Weights ...... 134 2.2.1 Review of Previous Applicable Studies and FRA Data ...... 136 2.3 Definition of Risk Factor Levels ...... 139 2.4 Assignment of Risk Factor Levels to Subdivisions ...... 139 3.0 Description of Risk Factors and Quantification of Risk Factor Levels and Weights ...... 140 3.1 Risk Factor #1: Annual Million Gross Ton (MGT) Level ...... 140 3.1.1 Risk Factor Overview ...... 140 3.1.2 Quantification of Risk Factor Weight ...... 141 3.1.3 Quantification of Risk Factor Levels ...... 141 3.2 Risk Factor #2: Presence and Volume of Passenger Traffic ...... 142 3.2.1 Risk Factor Overview ...... 142 3.2.2 Quantification of Risk Factor Weight ...... 143 3.2.3 Quantification of Risk Factor Levels ...... 143 3.3 Risk Factor #3: Presence and Volume of Toxic Inhalation Hazard / Poison Inhalation Hazard (TIH/PIH) Material (Loads and Residue) Transported ...... 144 3.3.1 Risk Factor Overview ...... 144 3.3.2 Quantification of Risk Factor Weight ...... 144 3.3.3 Quantification of Risk Factor Levels ...... 145 3.4 Risk Factor #4: Number of Tracks ...... 147 3.4.1 Risk Factor Overview ...... 147 3.4.2 Quantification of Risk Factor Weight ...... 148 3.4.3 Quantification of Risk Factor Levels ...... 148 3.5 Risk Factor #5: Method of Operation ...... 149 3.5.1 Risk Factor Overview ...... 149 3.5.2 Quantification of Risk Factor Weight ...... 152 3.5.3 Quantification of Risk Factor Levels ...... 152 3.6 Risk Factor #6: Speed of Train Operations ...... 153 3.6.1 Risk Factor Overview ...... 153 3.6.2 Quantification of Risk Factor Weight ...... 154 3.6.3 Quantification of Risk Factor Levels ...... 154 3.7 Risk Factor #7: Grade ...... 155 3.7.1 Risk Factor Overview ...... 155 3.7.2 Quantification of Risk Factor Weight ...... 155 3.7.3 Quantification of Risk Factor Levels ...... 155 3.8 Risk Factor #8: Curvature ...... 155 3.8.1 Risk Factor Overview ...... 156 3.8.2 Quantification of Risk Factor Weight ...... 156 3.8.3 Quantification of Risk Factor Levels ...... 157 3.9 Other Risk Factors Not Included in the Risk Prioritization Model ...... 157 4.0 Model Calculation Tool ...... 160 5.0 Risk Prioritization Model Results ...... 171

129

PTCIP

6.0 References ...... 172

List of Tables TABLE 2-1. IDENTIFICATION OF RISK FACTORS INCLUDED IN CN RISK PRIORITIZATION MODEL ...... 133

TABLE 3-1. FACTOR LEVELS FOR „ANNUAL MGT LEVEL‟ ...... 141 TABLE 3-2. FACTOR LEVELS FOR „PRESENCE AND VOLUME OF PASSENGER TRAFFIC‟ ...... 143 TABLE 3-3. FACTOR LEVELS FOR „PRESENCE AND VOLUME OF TIH/PIH MATERIALS‟ ...... 146 TABLE 3-4. FACTOR LEVELS FOR „NUMBER OF TRACKS‟ ...... 149 TABLE 3-5. FACTOR LEVELS FOR „METHOD OF OPERATION‟ ...... 153 TABLE 3-6. FACTOR LEVELS FOR „SPEED OF TRAIN OPERATIONS‟ ...... 154 TABLE 3-7. FACTOR LEVELS FOR „GRADE‟ ...... 155 TABLE 3-8. FACTOR LEVELS FOR „CURVATURE‟ ...... 157

List of Figures FIGURE 2-1. APPROXIMATIONS FOR CONSIDERATION IN ESTIMATION OF RISK FACTOR WEIGHTS .. 139

FIGURE A-1. APPROXIMATIONS FOR CONSIDERATION IN ESTIMATION OF RISK FACTOR WEIGHTS ...... 182

130

PTCIP

1. Introduction This document describes the risk prioritization model generated in response to 49CFR§236.1011(4), which requires that, to the extent practical, the positive train control (PTC) system be implemented to address areas of greater risk to the public and railroad employees before areas of lesser risk. The risk prioritization model assesses a number of key risk factors, which are assumed to provide an indication of the relative risk associated with the CN subdivisions for which PTC deployment is required by Subpart I §236.1005(b). The relative risk rankings generated by the risk prioritization model provided the basis for prioritizing deployment of PTC on the CN subdivisions for which PTC is required by Subpart I §236.1005(b). The risk prioritization model did not assess other CN subdivisions for which PTC deployment is not required by Subpart I §236.1005(b). This document describes the risk prioritization approach, the risk factors that were assessed, and the model results.

The risk factor prioritization model described in this document and used by CN was developed through a cooperative effort between a number of Class 1 Railways working with the Rail Safety group at Battelle.

131

PTCIP

2. Risk Prioritization Model Approach The risk prioritization model used by CN is based heavily on the sample methodology provided by the Federal Railroad Administration (FRA) in the Risk Prioritization Methodology for PTC System Implementation [2] (hereafter referred to as the Risk Prioritization Template). This is a basic weighted score approach in which the minimum critical risk factors identified in the Risk Prioritization Template and PTC Implementation Plan Template [3], were assigned integer scores, corresponding with level of risk, ranging from 0 (lowest risk) up to 5 (highest risk) for each of the CN subdivisions to be equipped with PTC. Each risk factor was also assigned a weight, which provided an indication of the “relative importance” of the factor in determining the overall risk ranking. Equation 1 below shows how, for n risk factors, a relative risk score was generated for each subdivision by multiplying the integer score assigned to the subdivision for a given factor (FRi) by the weight assigned to that factor (FWi), and summing the products of the n risk factors.

(Equation 1) Relative Risk Score for Subdivision =

In order to perform the above calculation, the following activities were undertaken: 1) Identify risk factors to be included in the risk prioritization model

2) Estimate risk factor weights (FWi) 3) Define the risk factor levels (from 0 to 5) that would be used to assign scores to the subdivisions for each risk factor

4) Assign integer scores (FRi) to each subdivision using the criteria defined in #3 above

Details of each of the activities listed above are provided in the subsections following.

2.1. Identification of Risk Factors The Risk Prioritization Template includes a list of seven risk factors, which it identifies as “minimum critical risk factors that must be addressed” in the risk prioritization model. These seven risk factors, which are listed below, correspond with the risk factors identified in §236.1011(a)(5) as minimum factors that shall be used to determine the sequence in which track segments will be equipped: 1. Annual million gross ton (MGT) levels 2. Presence and volume of passenger traffic 3. Presence and volume of TIH/PIH material (loads and residue) transported 4. Number of tracks 5. Method of operation 6. Speeds of train operations 7. Track grades and curvatures.

CN also considered whether additional risk factors, beyond those identified in the Risk Prioritization Template, should be considered for inclusion in the risk prioritization model. While other potential sources for risk were discussed, it was estimated that these other factors would have a negligible effect on risk relative to many of the other risk factors that had already been identified in §236.1011(a)(5) and the Risk Prioritization Template.

132

PTCIP

As shown in Table 2.1 below, a total of 8 discrete risk factors were ultimately included in the model. Table 2.1 contains 1) the risk factors included in the risk prioritization model, 2) the associated risk factor weights, as estimated when taking all of the risk factors in the far left column of the table into account,, and 3) the definition of the range values for each risk factor. The upper part of the table contains the risk factor data from the PTCIP Template while the lower part of the table contains the range values normalized for CN.

In Table 2.1, it can be seen that one of the risk factors identified was decomposed into separate factors. Rather than trying to create a combined „Track grades and curvatures‟ risk score for each subdivision and criteria for evaluating these scores, it seemed reasonable, and more transparent, to simply measure a „Grade‟ risk factor separate from a „Curvature‟ risk factor.

Table 2-1. Identification of Risk Factors Included in CN Risk Prioritization Model

CN Risk Factor Ranges Range Values Risk Factor Unit 0 1 2 3 4 5 MGT per year <5 MGT 5 - <15 MGT 15 - <25 MGT 25 - <35 MGT 35 - <45 MGT 45 MGT & over Passenger Trains trains / day 0 1 - 2 3 - 5 6 - 10 11 - 20 >20 TIH Volumes cars / year 0 1 - <100 100 - <1000 1,000 - <5,000 5,000 - <10,000 GT 10,000 Number of Tracks main tracks N/A GT 2 tracks N/A 2 tracks N/A 1 track Method of Ops N/A ATC / ACS TCS Bi-Dir ABS Dir ABS No Sig/TWC Train Speed max train speed mph N/A 0 - <15 15 - <30 30 - <60 60 - <80 80 & over Track Grade max. grade percent N/A LT 0.5% 0.5% - <1.0% 1.0% - <1.5% 1.5% - 2.0% 2.0% & over Track Curvature max curvature degrees N/A 0 - <2.0 2.0 - <4.0 4.0 - <6.0 6.0 - <8.0 8.0 & over Overall Priority weighted average 0 - <1.5 1.5 - <2.0 2.0 - <2.5 2.5 - <3.0 3.0 - <3.50 3.5 & over

Notes:

1) MGT: Use weighted average of MGT for subdivisions with multiple measurement sections 2) Passenger Trains: Any passenger traffic on subdivision establishes passenger ranking for entire subdivision 3) TIH Volume: Use maximum number of TIH cars (loads & residue) on subdivision if there are multiple measurement sections 4) Number of Tracks: Need to have more than 50% of route miles as double or multi-track to achieve lower risk rating for sub 5) Method of Operations: Need to have more than 50% of route miles as TCS or ABS to achieve lower risk rating for sub 6) Train Speed: Use maximum train speed on subdivision (passenger, intermodal or freight) 7) Track Grade: Use maximum grade on subdivision 8) Track Curvature: Use maximum curvature on subdivision 9) Overall Priority: Sum of values for all individual risk factors for a line segment.

133

PTCIP

2.2. Estimation of Risk Factor Weights One step in creating the relative risk score is quantifying the relative weight for each of the relevant risk factors. CN faced significant time constraints to complete this task both because FRA‟s final rule[1] was issued on January 12, 2010, and submission of the Positive Train Control Implementation Plan (PTCIP) was required by April 16, 2010, and also because these results were required early in the process of developing the PTCIP because they would heavily influence other planning-related components of the document. In addition, no quantitative information was identified that would be appropriate for direct application as risk factor weights. Accordingly, CN adopted a reasonable and practical approach to the task that relied heavily on expert judgment.

CN initially considered using only in-house experts for this task. Discussions among the North American Class I railroads resulted in an agreement to collaborate on these estimates, however, making additional experts available to augment CN‟s expertise. Although considerable weight was given to the input concerning general industry risks that was received from these outside experts, CN assigned final risk factor weights keeping in mind their specific applicability to CN‟s subdivisions.

Individuals from each of the seven North American Class I railroads with significant expertise in railroad operations were considered the most qualified to provide the necessary judgments. Their detailed understanding of how the railroads operate and their familiarity with the various risks involved in railroad operations qualified these individuals to provide expert judgments.

The possibility of performing formal individual elicitations of these experts, using an elicitation approach based on the approach outlined in US Nuclear Regulatory Commission Regulation 1150 (NUREG-1150) [4] was considered, but it was decided that such an approach was not ideally-suited for quantification of these risk factor weights. This was primarily due to the fact that the risk factor weights are “high level” and rather abstract quantities, whereas formal elicitations are generally more effective when eliciting quantities at a “lower level” (for example, asking experts to quantify the probability that a train operator fails to obey a restrictive signal indication may be an estimate that is more effectively elicited in a formal elicitation setting). Also, while those deemed most suitable to estimate risk factor weights were personnel from the railroads with significant expertise in railroad operations, these experts were not necessarily experts in risk assessment, so it was recognized that additional collaboration among railroad operations experts and risk assessment experts would be beneficial. Lastly, time limitations also made a more formal elicitation process impractical. Accordingly, a more collaborative/group- oriented approach was adopted, with experts in risk assessment interacting with railroad operations experts in estimating risk factor weights.

Although a formal elicitation approach was not followed, key aspects of that approach were implemented. The experts, for example, were briefed on a number of common pitfalls that may be encountered when providing expert judgments, including susceptibility to various types of biases and also weaknesses related to providing estimates in a group setting.

Among the primary biases of concern was the “anchoring” bias, which refers to the tendency of starting with an initial estimate and failing to move from that estimate. In the case of risk factor

134

PTCIP weights, a potential source of anchoring bias was the Risk Prioritization Template; the experts could have potentially been biased by the weighting values included in this document even though the FRA had indicated that the values in the template were for illustrative purposes only. The experts were also made aware of the “availability” bias, which is the tendency to give too much weight to easily accessed or remembered information. An event receiving much attention often may result in a similar event being given a higher probability of occurring. For example, it is possible that one or more of the risk factors could have contributed significantly to a recent accident that is fresh in the minds of the experts, thereby creating the potential for the experts to give extra weight to that factor. The experts were cautioned that, while all relevant information should be taken into consideration, they should take care not to give too much weight to something due to the fact that it stands out in their minds simply because it was a recent occurrence.

An additional bias that was identified as a particular concern in performing an activity such as this is the “motivational” bias. This bias involves consciously providing judgments that do not reflect the expert‟s true beliefs. This can occur if the expert has an economic, political or other stake in the results of the assessment. This is not to suggest that any of the experts involved in this estimation exercise expressed this bias, it was a topic that was addressed so as to make the experts aware that this type of bias because one of the most effective ways to eliminate biases is to be aware of their potential to exist.

Other topics discussed with the experts before estimating risk factor weights were the advantages and disadvantages of the group discussions that took place during this process. In general, group discussions provide an opportunity to take advantage of the collective experience of the individual group members. The group setting can also lead to significant synergy, as ideas of the group may form a greater whole than the sum of individual ideas. But it also comes with some disadvantages, and the best way to address and minimize the disadvantages of group discussion is to be aware of their potential to exist. These include the potential for discussions to be dominated by one or a group members, the potential for group members to bias other group members, the potential for polarized opinions to create extra complexity, and the potential for the group to trend towards a compromised opinion. Detailed discussions of these and other related topics are widely available in a number of sources.

Another key topic that was discussed with the railroad operations experts, which is important to note here, is the fact that there would likely be a significant amount of uncertainty in their estimates, meaning that there would be significant uncertainty in the final risk factor weights, and ultimately a great deal of uncertainty in the priority with which CN subdivisions should be equipped with PTC. That uncertainty was not quantified. We note, however, that regardless of the approach that might have been taken to establish the risk factor weights there would have been a great deal of uncertainty.

The Risk Prioritization Template provided by FRA offered guidelines for assessing the prioritization with which segments should be equipped, and it was assumed that these guidelines provided a reasonable indication of the level of detail that the FRA was expecting to see in the railroads‟ risk prioritization models. Since the results of the risk prioritization are presented as

135

PTCIP risk scores, with no error bars or similar, it is important for those using the results in their decision-making process (in this case, personnel at CN who will decide the sequence with which subdivisions should be equipped) to understand that there is a significant amount of uncertainty in these results. The results should be interpreted as “best estimates” of rail operations experts.

To help inform and validate the expert‟s judgments, and to reduce the epistemic uncertainty of their estimates, a review of the pertinent literature was conducted. This information gathering involved the activities described below. 1) Review of previous applicable studies and existing FRA data: An effort was made to identify and review previous studies whose findings could be taken into consideration when estimating the risk factor weights to be used in the risk prioritization model. A summary of these reviews is included in Subsection 2.2.1, with more detailed discussion of them provided in Appendix C. A number of other sources of FRA data were also reviewed as part of an effort to identify key data that should be taken into consideration when estimating risk factor weights. Due to the fact that the majority of available FRA data is not specific to PTC-preventable accidents, it was determined that it had limited value given time constraints. If additional time were available, FRA data could have been reviewed for its applicability to PTC (similar to the identification of PTC-preventable accidents performed by the Accident Review Team documented in the 1999 RSAC report [5]). 2) Written description of risk factors: The results of this collaborative effort of operations experts from the seven North American Class I railroads are included in Section 3.0 below. Findings from the studies and other data reviewed above were cited or included in the discussion of the individual risk factors, where applicable.

After these information-gathering activities were completed, the experts were asked to provide quantitative estimates, taking the compiled information into consideration. Since the experts agreed that their respective railroads may want to consider different subsets of the risk factors identified in Table 2.1 in their risk prioritization models, experts were asked to provide estimates for the seven “minimum critical risk factors,” and railroads then used those estimates as they saw fit for their own assessment and considered additional risk factors beyond this minimum set, as appropriate for their particular circumstances.

2.2.1. Review of Previous Applicable Studies and FRA Data This section briefly summarizes reviews of previous research related to the railroad risk factors that were included in the risk prioritization model. The three key references that were reviewed are identified in the bulleted list below. In addition to considering information from these references, reviews of various other railroad-related reports and FRA data was also performed. Any relevant information identified in these other sources has been included in other sections of this document, along with citations as appropriate, but sources beyond the three listed below are not addressed in this subsection.  Railroad Safety Advisory Committee (RSAC), Federal Railroad Administration, US Department of Transportation. Report of the Railroad Safety Advisory Committee to the Federal Railroad Administrator: Implementation of Positive Train Control Systems. September 1999. [5]

136

PTCIP

 Railroad Systems Division (DTS-75), Office of Safety and Security, The John Volpe National Transportation Systems Center, US Department of Transportation. Presentation for Office of Safety, Federal Railroad Administration RSAC/PTC Working Group Risk 2 Team, Base Case Risk Assessment: Data Analysis & Tests. April 22, 2003. [6]  Carl D. Martland, Ying Zhu, Youssef Lahrech, and Joseph M. Sussman. Risk and Train Control: A Framework for Analysis. Center for Transportation Studies, Massachusetts Institute of Technology, Cambridge, January 2001. [7]

For each of the above-cited references, the subsections following provide a brief overview of the contents of each reference and a discussion of key results/findings that were considered when assigning risk factor weights for use in the risk prioritization model. Appendix C contains additional discussion of each of these references.

2.2.1.1. Report of the Railroad Safety Advisory Committee: Implementation of Positive Train Control Systems The RSAC Report of the Railroad Safety Advisory Committee to the Federal Railroad Administrator: Implementation of Positive Train Control Systems [5] made use of historical data to help estimate PTC benefits by evaluating PTC-preventable accidents (PPAs). RSAC identified 819 PPAs out of the more than 25,000 accidents reported to FRA between 1988 and 1997 and attempted to estimate the benefits of PTC by evaluating how many accidents would have been eliminated were a given type of PTC system in place. These data were input into the Corridor Risk Assessment Model (CRAM) model, which was used to estimate the safety benefits of PTC by relating the historic occurrence and consequences of accidents that may have been prevented by a PTC system to specific track features and traffic.

Some of the details of the regression analyses that led to the findings/results identified in Appendix C are not included in the RSAC report, and therefore, the findings/results were considered with some caution. There is nothing, however, to indicate that the analysis contained in this report is not valid or useful.

In general, the conclusions of the report indicate that the „Annual MGT Level‟ risk factor should potentially be assigned a relatively high weight. The „Curvature‟ risk factor may not be a very strong indicator of risk, so it should potentially be assigned a relatively low weight. „Speed of Train Operations‟ actually had a counter-intuitive effect on predicting the occurrence of accidents, but this may not be an unreasonable conclusion due to the fact that authorized speeds may be lower/higher on a section of track depending on the presence/absence of other risk factors. The report also provides information that helps illustrate the noticeable effect on risk that „Presence and Volume of TIH/PIH Material‟ and „Presence and Volume of Other Non- TIH/PIH Material‟ could potentially have, indicating that one or both of these factors should be weighted fairly high.

2.2.1.2. Base Case Risk Assessment: Data Analysis & Tests (Volpe Center) The Base Case Risk Assessment presentation [6] describes work performed by the Volpe Center in summarizing the “current level of risk” for a number of territories, as it summarizes results from various

137

PTCIP linear and anti-log multiple regression analyses that were performed. The study examines segments where PPAs have taken place, as well as other mainline railroad segments. The analyses described in this presentation are assumed to be a follow-on to the analyses documented in the RSAC report summarized in the previous subsection.

For reasons discussed in additional detail in Appendix C, most of the conclusions of this presentation were not taken into consideration when assigning risk factor weights. However, data that were taken into consideration include data on annual PPA cost/train-mile, which were estimated to be 0.0544 for “Auto” (ATS or cab signaled) territory, 0.0675 for TCS territory, 0.0590 for ABS territory, and 0.122 for dark territory. These data indicate that risk per train mile in dark territory is approximately two times the risk of other territories.

2.2.1.3. Risk and Train Control: A Framework for Analysis The report Risk and Train Control: A Framework for Analysis by Martland et al. [7] documents a PRA study which examined the effects of train control strategies on the risks of railroad operations. The research was initiated to address the question of how risk and the potential for risk reduction varies from one rail corridor to another, and it also attempts to assess how much reduction in risk can be expected from different train control strategies. The study considers a hypothetical corridor, which the report indicates was similar to the 183 actual corridors analyzed in the RSAC Report of the Railroad Safety Advisory Committee to the Federal Railroad Administrator: Implementation of Positive Train Control Systems [5], discussed in Subsection 2.2.1.1 above.

While this study seems to provide an indication of the extent to which different risk factors affect total risk under different methods of operation, it is acknowledged that various aspects of the CN subdivisions that are subject to the risk prioritization model vary from the segments that compose the hypothetical corridor assessed in this study. CN was mindful of differences such as these when considering these results during their estimation of risk factor weights. It was also acknowledged that the PRA and sensitivity analysis results presented in this study should only be considered directly applicable within the variable ranges assessed for the hypothetical corridor, such that extrapolation beyond these ranges is generally not recommended.

Keeping the above caveats in mind, it is also true that the results of this PRA study are among the few available data points that are somewhat applicable to the risk factor weighting exercise, and while it may not be advisable to give too much weight to the results of this study, it did seem reasonable to consider how the risk factors rank with regard to the extent to which they affect risk, and it also seemed reasonable to consider the results as “rough order of magnitude” estimates to be considered when assigning risk factor weights.

The approximations discussed in Appendix C were made in order to arrive at rough order of magnitude estimates for risk factor weights. Figure 2-1 below shows the relative weights of the four risk factors that were approximated. These approximations are also briefly discussed in Section 3.0 under the various risk factors to which these estimates are applicable. The approximations were just one set of inputs to be considered when assigning risk factor weights and were interpreted simply as rough approximations that were made using available information.

138

PTCIP

Figure 2-1. Approximations for Consideration in Estimation of Risk Factor Weights

2.3. Definition of Risk Factor Levels A discussion of each of the risk factors identified in Table 2.1 above is provided in Section 3.0 below. Included with the discussions of each risk factor in Section 3.0 are the definitions of the various scoring levels (from 0 to 5) that were used to rate the subdivisions for each risk factor. In general, risk factor levels were defined by looking at the range of data for a risk factor across all subdivisions and assigning factor levels that were typically evenly-spread between the minimum and maximum values of the data. In some cases, levels were slightly adjusted for the purposes of grouping similar subdivisions (with respect to a particular risk factor) within the same factor scoring level.

The factor scoring levels are typically provided in small tables towards the end of the discussion of each risk factor in Section 3.0. The upper and lower limits of each scoring level are defined in these tables. Where applicable, plots of available CN risk factor data, which show the distribution of subdivisions among the factor levels defined for each risk factor, are also provided.

2.4. Assignment of Risk Factor Levels to Subdivisions The data that were used to quantify the various risk factors for each of the CN subdivisions included in the risk prioritization model were collected from the appropriate departments within CN. . Data were converted, as necessary, to be compatible with the factor levels that were defined for each risk factor so that the CN subdivisions could be assigned a score (also referred to as „Risk Factor Levels‟) for each of the risk factors. Section 3.0 below provides additional information on the risk factor levels that were defined for each risk factor and describes how the data were used to quantify the various risk factors included in the risk prioritization model. The actual risk factor scores assigned to each subdivision for each risk factor are provided in Table 4.1 in Section 4.0 later in this report.

139

PTCIP

3. Description of Risk Factors and Quantification of Risk Factor Levels and Weights The various risk factors that were considered for use in CN‟s PTCIP risk prioritization model are described in the subsections following. A brief discussion of how each factor is generally assumed to affect the relative risk on CN subdivisions and how this risk may be affected by implementation of PTC is provided under the Risk Factor Overview heading for each risk factor. Also, under each risk factor, the Quantification of Risk Factor Weight subsection discusses the weight assigned to the risk factor, and the rationale for assigning the weight. The Quantification of Risk Factor Levels subsection under each risk factor describes factor levels that were used to “score” each of the subdivisions. Also provided for factors that were included in the model is a listing of the factor levels that were defined during the quantification activities and plots of available data. Table 2-1 provided in Section 2 contains a summary of the assigned risk factor levels and weights for each risk factor. The spreadsheet tool that was used to score each subdivision for each risk factor, and which was also used to perform the calculations shown in Equation 1 above, is described in Section 4.0.

The risk factors described in Subsections 3.1 through 3.8 are the risk factors identified in §236.1011(a)(5) as minimum factors that shall be used to determine the sequence in which track segments will be equipped.

In the descriptions of risk factors included throughout this section, there are a number of statements that address how different risk factors affect risk of rail operations. It should be understood that these statements are being made assuming that “all other factors remain equal.” This is being noted here so as to avoid excessive repetition of statements such as “All other things being equal…”, or similar, throughout the descriptions.

3.1. Risk Factor #1: Annual Million Gross Ton (MGT) Level 3.1.1. Risk Factor Overview This risk factor provides an indication of the annual volume of traffic present on a subdivision. An increased volume of traffic corresponds with increased probability of various types of accidents. This is particularly the case on single track segments, where meets and passes increase as total train traffic increases, creating an increased probability of train-to-train collision. On segments of multiple track, meets and passes are not increased to the same extent due to increased capacity. As volume of traffic increases, established worked zones would be expected to be encountered by more trains, creating greater opportunity for incursion into work zones. Any sections of track that might provide greater opportunity for overspeed derailment (areas of extreme grade or curvature, for example) would be traveled by more trains, thereby creating increased opportunity for derailment. Increased operation of switches required to accommodate a higher volume of traffic may allow greater opportunity for switches to be misaligned. Also, while increased volume would primarily affect the probability component of risk, it is expected that consequences could be slightly increased on subdivisions with higher traffic volume due to the fact that the occurrence of an accident could require re-routing of an increased number of trains and cause greater disruption to commerce than an accident on a less-traveled subdivision, thereby resulting in increased economic costs.

Since the potential for these various conditions to occur is assumed to increase along with the MGT level, the implementation of PTC would be expected to achieve greater reductions in risk for subdivisions with higher MGT levels. The installation of PTC would be expected to decrease the risk associated with train- to-train collisions, overspeed derailments, incursions into established work zone limits, and mishaps resulting from movements through improperly-positioned main line switches.

140

PTCIP

3.1.2. Quantification of Risk Factor Weight Taking into consideration the information discussed in the „Overview‟ section above, and considering how this risk factor compares with the other risk factors described in Section 3.0, it seemed clear that this risk factor should receive a moderate-to-high weight. While the findings of the RSAC Report on Implementation of Positive Train Control Systems [5] are being considered with caution, this report concluded that, of the corridors studied, the highest predictor of risk was the volume of traffic.

When estimating the weighting value of this risk factor, consideration was also given to the fact that it is one of three risk factors, along with „Volume of PIH/TIH Traffic‟ and „Presence and Volume of Passenger Traffic‟ that serve as the primary criteria for determining which lines are to be equipped with PTC, as required by §236.1005(b)(1). The fact that Congress used these three factors in their criteria in the Rail Safety Improvement Act of 2008 [8] would seem to indicate that all three are significant factors to consider when assessing the risk of railroad operations.

CN assessed this risk factor to have a “very high” effect on the probability component of risk and a “medium” effect on the consequence component of risk. The other railroads generally seemed to draw similar assessments. As a result, the total risk weight assigned to this risk factor was 21%. As can be seen in Table 2.1, this weight is essentially the same as that assigned to Passenger Traffic and TIH/PIH Levels. 3.1.3. Quantification of Risk Factor Levels For each subdivision with multiple MGT measurement sections, a weighted average of the MGT levels for each measurement section was used to generate an aggregate measurement for the subdivision (Weighted Avg MGT = Sum of (section MGT x section miles) / Subdivision Miles).

Limits for six factor levels, provided in Table 3-1 below, were assigned taking the range of CN subdivision MGT data into account.

Table 3-1. Factor Levels for ‘Annual MGT Level’ Factor Level Lower Limit Upper Limit Level 0 0 MGT <5 MGT Level 1 5 MGT <15 MGT Level 2 15 MGT <25 MGT Level 3 25 MGT <35 MGT Level 4 35 MGT <45 MGT Level 5 45 MGT None

141

PTCIP

3.2. Risk Factor #2: Presence and Volume of Passenger Traffic 3.2.1. Risk Factor Overview This risk factor addresses the volume of passenger trains that travel on CN subdivisions on a daily basis. Due to the fact that significant value is assigned to injury and loss of life when assessing the consequences of an accident, accidents involving passenger trains generally result in increased consequences relative to accidents involving only freight trains, for example. Serious accidents involving freight trains are generally not expected to result in injury or death in excess of the railroad operations personnel on-board each train, while accidents involving passenger trains could potentially result in injury or death of operations personnel, as well as the injury or death of passengers aboard the train.

While this factor is primarily viewed as a contributor to the consequence component of risk, the probability of an accident would also increase as the volume of passenger traffic increases on a subdivision. Passenger traffic volumes are not taken into account in the „Annual MGT Level‟ risk factor discussed above, so when assessing the amount of traffic on a subdivision, passenger traffic volume should be considered in addition to MGT values, thereby acting to increase subdivision risk from train volume for reasons previously described in the description of the „Annual MGT Level‟ risk factor above. In addition to the increases in risk described above, the presence of passenger traffic may also have a significant effect on traffic flow. Passenger traffic is, in some cases, given movement priority, which can increase the number of passes directly. This, in turn, may lead to more meets between freight trains and increased risk from the need to operate and travel across switches.

PTC may serve to reduce the risk associated with passenger trains via efficiency improvements and enforcement (preventing collisions and derailments). The installation of PTC would generally be expected to decrease the risk associated with train-to-train collisions, overspeed derailments, and mishaps resulting from movements through improperly-positioned main line switches, all of which are accident types that could involve passenger trains. Since the potential for accidents involving passenger trains would clearly increase along with passenger traffic, the installation of PTC would be expected to achieve greater reductions in risk for subdivisions with higher volumes of passenger traffic.

The risk associated with passenger train stations, which are required to accommodate passenger traffic, is also assumed to increase with higher volumes of passenger traffic, as higher volumes of passenger traffic would generally be expected to coincide with greater numbers of individuals congregating at stations, waiting for, boarding, and disembarking from passenger trains. Stations are typically areas of congregation which may, at times, have high concentrations of people located relatively close to the track. Accidents occurring at or near passenger stations, particularly accidents involving TIH/PIH or other hazardous materials, could potentially have significantly increased consequences relative to accidents occurring on other portions of the track, away from passenger stations. As discussed above, implementation of PTC would be expected to reduce the likelihood of various types of accidents, and this includes accidents at or near passenger stations.

142

PTCIP

3.2.2. Quantification of Risk Factor Weight Taking into consideration the information discussed in the „Overview‟ section above, and considering how this risk factor compares with the other risk factors described in Section 3.0, it seemed clear that this risk factor should receive a relatively high weight. The report Risk and Train Control: A Framework for Analysis [7] discussed in Section 2.2.1.3 concluded that “Passenger traffic is the most important factor because the addition of passenger trains creates the possibility of catastrophic accidents with dozens of fatalities,” however, it should again be noted that the report referenced above did not study other factors that FRA has identified as “minimum critical factors that must be addressed,” such as „Method of Operation‟ and „Volume of PIH/TIH Traffic‟.

The FTA‟s 2006 Commuter Rail Safety Study [11] indicated that, between January 1, 2001 and July 31, 2006, a total of 15 commuter rail passengers and three commuter rail employees were killed in collisions or derailments (the number of these fatalities that were PTC-preventable was not identified), indicating that, although the potential for serious consequences is very real and the occurrence of even a single fatality is significant, the consequences of accidents involving passenger traffic over that period of time did not reach the “dozens” suggested possible in the „Overview‟ section above. However, the September 12, 2008 accident at Chatsworth, California does provide an illustration of how severe the consequences can be, as 25 people were killed and over 130 more were seriously injured. Although NTSB has not yet released its final report, evidence summarized at the NTSB‟s public hearing suggested that this is the type of accident that potentially could have been prevented by PTC. The occurrence of several other accidents, inside the United States and in other developed European nations with comparable infrastructure, have resulted in significant consequences of this magnitude and greater. As was noted under the „Annual MGT Level‟ risk factor above, consideration was also given to the fact that this is one of three risk factors, along with „Annual MGT Level‟ and „Volume of PIH/TIH Traffic‟ that serve as the primary criteria for determining which lines are to be equipped with PTC, as required by §236.1005(b)(1). The fact that Congress used these three factors in their criteria in the Rail Safety Improvement Act of 2008 [8] would seem to indicate that they are significant factors to consider when assessing the risk of railroad operations.

CN assessed this risk factor to have a “medium” effect on the probability component of risk and a “very high” effect on the consequence component of risk. The other railroads generally seemed to draw similar assessments. As a result, the total risk weight assigned to this risk factor was 21%. As can be seen in Table 2.1, this weight is essentially the same as that assigned to Annual MGT and TIH/PIH Levels. 3.2.3. Quantification of Risk Factor Levels For each subdivision, the highest number of passenger trains per day running on one or more of the subdivision segments determined the passenger train level for the entire subdivision. Data is based on 2008 passenger train volumes. Limits for six factor levels, provided in Table 3-2 below, were assigned taking the range of passenger train volume data into account.

Table 3-2. Factor Levels for ‘Presence and Volume of Passenger Traffic’ Factor Level Lower Limit Upper Limit Level 0 0 Trains/Day 0 Trains/Day

143

PTCIP

Level 1 1 Trains/Day 2 Trains/Day Level 2 3 Trains/Day 5 Trains/Day Level 3 6 Trains/Day 10 Trains/Day Level 4 11 Trains/Day 20 Trains/Day

Level 5 21 Trains/Day None

3.3. Risk Factor #3: Presence and Volume of Toxic Inhalation Hazard / Poison Inhalation Hazard (TIH/PIH) Material (Loads and Residue) Transported 3.3.1. Risk Factor Overview This risk factor provides an indication of the volume of TIH/PIH materials that are transported on a subdivision on an annual basis. The presence of TIH/PIH materials could significantly increase the consequences of an accident, as accidents involving trains carrying TIH/PIH materials may not only cause injury and/or loss of life to those onboard the train and in the railroad right-of-way, but also injury and/or loss of life to the general population in areas beyond the railroad, potentially even miles away from the location of the accident. Depending on the physical nature of the accident (extent of damage to cars, specific materials, chemical phase, flow rates, meteorology, population density in the affected area, etc.), a TIH/PIH release could potentially cause physical harm to large numbers of exposed individuals. Accidents involving TIH/PIH materials could also result in significant damage to the environment and other economic costs from activities such as accident remediation, evacuation, treatment of the “worried well”, loss of domesticated livestock, potential re-routing of rail and highway traffic, etc. This risk factor primarily affects the consequence component of risk for the reasons described above, and due to the fact that TIH/PIH loads are accounted for in the „Annual MGT Level‟ risk factor (i.e., this risk factor does not represent the presence of additional traffic), it is estimated that the „Presence and Volume of TIH/PIH Material‟ risk factor does not significantly affect the probability component of risk in this assessment.

The installation of PTC would generally be expected to decrease the risk associated with train-to- train collisions, overspeed derailments, and mishaps resulting from movements through improperly-positioned main line switches, all of which are accident types that could involve trains transporting TIH/PIH loads and residue. Also, by preventing incursions into established work zone limits, some additional reduction in risk specific to TIH/PIH materials may be achieved. Since the potential for accidents involving TIH/PIH materials would clearly increase along with the volume of TIH/PIH materials transported, the implementation of PTC would be expected to achieve greater reductions in risk for subdivisions upon which higher volumes of TIH/PIH loads and residues are transported. 3.3.2. Quantification of Risk Factor Weight Taking into consideration the information discussed in the „Overview‟ section above, and considering how this risk factor compares with the other risk factors described in Section 3.0, it seemed clear that this risk factor should receive a relatively high weight. While this risk factor was not quantitatively assessed in the three reports that were discussed in Section 2.2.1, the RSAC Report on Implementation of Positive Train Control Systems [5], in referring to the 1988 – 1997 accident data that were analyzed, notes that “The trends in the derailment category indicate relatively infrequent low-consequences events, whose greatest potential hazard is in the possibility of the release of hazardous chemicals requiring an evacuation. Seventeen of four- hundred twenty derailments resulted in evacuations; the average number of people evacuated

144

PTCIP was approximately 420 per incident. Two incidents resulted in over 1000 evacuations. One derailment, included in the group of accidents thought to be possibly preventable by the highest level of PTC system1, accounted for 47 fatalities. This accident is not consistent with the general trend of the consequences of PTC-preventable derailments being less than collisions, but it identifies a source of risk.”

Potential consequences from accidents involving TIH/PIH materials could be orders of magnitude higher than the accidents referenced in the above report; however, the above statements do help illustrate the low-probability, high-consequence potential of this risk factor. Although the report Risk and Train Control: A Framework for Analysis [7] also did not directly assess this risk factor, it noted that “Accidents involving hazardous materials could conceivably be much more catastrophic even than a high-speed collision of two passenger trains,” and it also stated that “Conceivably, the potential for a catastrophic PPA involving hazardous materials would add significantly to the risks identified in this paper.”

As noted for the two risk factors previously addressed in Section 3.0, consideration was given to the fact that „Presence and Volume of TIH/PIH Material‟ is part of the primary criteria for determining which lines are to be equipped with PTC, as required by §236.1005(b)(1), which indicates that this is a factor that should figure heavily into the assessment of the priority with which subdivisions should be equipped with PTC. One additional item of note from the preamble of Part 236 Subpart I, however, is FRA‟s statement, on page 2623, that “If FRA were writing on a blank slate, the agency may have considered factors that drive risk and thresholds for those factors, taking into consideration more than PIH and intercity or commuter passenger traffic. Some lines that the Congress has required to be equipped by the end of 2015 because of PIH traffic would be left for deployment well downstream. Under such a hypothetical scenario, others with heavy train counts or without signal systems (and with robust traffic) may have been in theory added to the list for deployment of PTC by the end of 2015.” Although this statement does not discount the risk associated with PIH/TIH traffic, it does seem to indicate that factors such as „Annual MGT Level‟ and „Method of Operation‟ may warrant equal/similar weighting.

CN assessed this risk factor to have a “Low” effect on the probability component of risk and a “very high” effect on the consequence component of risk. The other railroads generally seemed to draw similar assessments. As a result, the total risk weight assigned to this risk factor was 20%. As can be seen in Table 2.1, this weight is essentially the same as that assigned to Annual MGT and Passenger Traffic Levels 3.3.3. Quantification of Risk Factor Levels For subdivisions where TIH/PIH materials were transported in 2008, the number of cars per year which transport TIH/PIH materials were identified. For those subsections with multiple measurement sections, the maximum number of TIH cars for one or more of the sections was used for the entire subsection.

1 As a part of the RSAC process, the accident review team analyzed accident data to determine which accidents might have been preventable by each of four different PTC design concepts. The four design concepts are hierarchical in that “higher” concepts incorporate all functions and coverage of any “lower” concept(s). The “highest level of PTC system” referenced in this quotation refers to the PTC design concept at the top of this hierarchy; that is, the concept that incorporates the most functionality and coverage. Details of this “highest level of PTC system” can be found in the referenced report.

145

PTCIP

Limits for six factor levels, provided in Table 3-3 below, were assigned taking the range of CN subdivision TIH/PIH data into account.

Table 3-3. Factor Levels for ‘Presence and Volume of TIH/PIH Materials’

Factor Level Lower Limit Upper Limit Level 0 0 Cars/Year 0 Cars/Year

Level 1 1 Cars/Year <100 Cars/Year Level 2 100 Cars/Year <1,000 Cars/Year Level 3 1,000 Cars/Year <5,000 Cars/Year Level 4 5,000 Cars/Year <10,000 Cars/Year

Level 5 10,000 Cars/Year None

146

PTCIP

3.4. Risk Factor #4: Number of Tracks 3.4.1. Risk Factor Overview This risk factor addresses the presence of multiple main line tracks along portions of a subdivision. The increased capacity provided by multiple tracks allows for the number of train meets and passes to be reduced. A reduction in the number of meets and passes would be expected to decrease the number of train-to-train collisions, such that the presence of multiple tracks provides a reduction in risk relative to instances of single track. Although traffic volume may tend to be increased on segments with additional capacity provided by multiple tracks, it is estimated that an increase in traffic volume would not be expected to offset the benefits provided by multiple tracks. As the „Number of Tracks‟ risk factor is therefore assumed to affect the frequency with which collisions occur, it primarily affects the probability component of risk.

Although the presence of multiple tracks is generally associated with reduced risk, a slight increase in risk likely results from the fact that the presence of trains on adjacent tracks provides the opportunity for secondary collision, in which a derailed train may move onto an adjacent track, striking another train (note that this risk may also exist in areas of single track with features such as sidings). The presence of multiple track also provides increased opportunity for broken rail conditions, simply due to the fact that there is more track available for this condition to occur. Although this may be true, the vast majority of segments with multiple track are also signaled, which greatly reduces the potential for a broken rail to lead to an accident. One additional “negative” aspect of multiple track is the fact that it is generally considered less safe than single track when falling within a work zone, as increased awareness of workers and train operators is required.

To some extent, the disadvantages of multiple track identified above reduce the risk differential between single track and multiple track. Since this risk prioritization model is concerned with implementation of PTC, it was necessary to consider how these various increases and decreases in risk would be affected by installation of PTC. It is expected that PTC will prevent many of the train-to-train collisions that might have otherwise taken place on segments of single track through warning and enforcement of the limit of authority, while not preventing as many collisions on double track simply due to the fact that not as many collisions would have been expected to occur on double track. As a result, installing PTC on single track would be expected to have increased benefit in this respect.

Although PTC cannot prevent a train that has derailed from an adjacent track from colliding with a train on a PTC-equipped line, PTC is expected to reduce the number of derailments, such that the initial derailment which might have caused a secondary collision never takes place, thereby reducing the risk of secondary collisions. This would indicate that some additional reduction in risk may be achieved on segments of multiple track that would not have been achieved on single track. In addition, some slight benefit to installing PTC on segments of multiple track relative to single track would be achieved in signaled territory due to the added enforcement of restrictions in advance of broken rail conditions on these segments (increased benefit on multiple track since increased opportunity for broken rail to occur). Since it was assumed that risk associated with broken rail conditions would not be reduced by PTC in non-signaled territory, no additional benefit would be provided to multi-track in non-signaled territory. One final way in which

147

PTCIP implementation of PTC on multi-track territory may provide greater benefit in terms of risk reduction relative to single track is in the form of reducing the risk in work zones. Page 2646 of the preamble to Part 236 Subpart I supports this claim, as it states that “Any place on the national rail system is a potential roadway work zone, but special challenges are presented in providing for on-track safety where train movements are very frequent or operations are conducted on adjacent tracks.” It is expected that implementation of PTC might provide at least a slight additional benefit in protecting work zones in areas of multiple track relative to single track.

While installing PTC on segments of multiple track may have some benefits relative to single track, these benefits would likely be trumped by the anticipated reduction in train-to-train collisions on single track. However, the additional risk reductions achieved by installing PTC on segments of single track appear to be less significant than MGT, passenger train, or TIH levels, and this was ultimately reflected in the weight assigned to the „Number of Tracks‟ risk factor. 3.4.2. Quantification of Risk Factor Weight Taking into consideration the information discussed in the „Overview‟ section above, and considering how this risk factor compares with the other risk factors described in Section 3.0, it seemed clear that this risk factor should receive a relatively low weight compared to some of the other “minimum critical” risk factors. The report Risk and Train Control: A Framework for Analysis [7] concludes that “Single-track operations are much more susceptible to collisions …,” and that “The potential benefits of PTC systems also vary with the amount of multiple tracks. All three systems (referring to three PTC technologies) will be slightly more effective in single- track than in multiple-track situations.” The rough approximations that were made using data from that report, as described in Section 2.2.1.3 above, indicated that it was the least important of the four primary risk factors studied in the sensitivity analyses. The weight approximated for „Presence and Volume of Passenger Traffic‟ was roughly an order of magnitude greater than the weight approximated for „Number of Tracks‟. „Annual MGT Level‟ was five times that of „Number of Tracks‟, and „Speed of Operations‟ was over three times that of „Number of Tracks‟. One additional item to note, related to the study documented in Risk and Train Control: A Framework for Analysis, is that the study did not appear to take into account some of the risk- related aspects described in the „Overview‟ section above, including the issue of work zones and secondary collisions. As a result, the reduction in risk achieved in single track territory relative to multiple track territory could potentially be slightly overstated.

3.4.3. Quantification of Risk Factor Levels Track data and maps were used to identify instances of multiple main lines within a subdivision. The mileage for each of these instances of multiple track was then identified and the total distance of multiple track was found for each subdivision. The percentage of double or multi- track (greater than 2 tracks) on a subdivision was then calculated using the distance of multiple track and the total distance of the subdivision to be equipped.

148

PTCIP

Since overall risk should decrease with the number of tracks, a subdivision needed to have more than 50% of its route miles as double or multi-track to be classified at the lower risk rating. That is, a subdivision needed to have at least 51% of its route miles as double track to be classified as double track and more than 51% of its route miles as multi- track to be classified as multi-track track.

Limits for three factor levels, provided in Table 3-4 below, were defined at the one track, two track, and greater than 2 track levels.

Table 3-4. Factor Levels for ‘Number of Tracks’ Factor Level Lower Limit Upper Limit Level 1 > 2 Tracks None Level 3 2 Tracks 2 Tracks Level 5 1 Track 1 Track

3.5. Risk Factor #5: Method of Operation 3.5.1. Risk Factor Overview The „Method of Operation‟ risk factor addresses the method by which trains are operated on a subdivision. This risk factor primarily affects the probability component of risk. Different methods of operation may offer varying levels of reliability/safety in coordinating the movement of trains and in detecting and responding to unsafe conditions that may exist in the operating environment, thereby affecting the frequency with which accidents occur. In general, PTC is expected to provide greater risk reduction on subdivisions where the method of operation currently offers lower levels of reliability/safety. While the implementation of PTC might be expected to substantially decrease the frequency of accidents, and therefore risk, on subdivisions that employ a particular method of operation, it may result in only modest decreases in accident frequency and risk for territories operated under other methods of operation.

CN controls all of their subdivisions using one or more of the three general methods of operation identified below (some subdivisions employ more than a single method of operation). A brief description of each method of operation and comments on the relative risk associated with each method of operation, in general and in the context of PTC implementation, are also provided. 1. Non-Signaled Track Warrant Control (TWC): Non-signaled TWC is a method of operation that uses written authorizations that are verbally communicated to control train movements on a main track within specified limits in a territory designated by the timetable. The track warrant permits a specific train to occupy a specific piece of track between named locations. The maximum allowable speeds within these non- signaled/dark territories are 49 mph for freight trains and 59 mph for passenger trains, subject to civil speed limits. Non-signaled TWC territories also provide greater opportunity for misaligned switches, fouling equipment, and broken rails to lead to an accident due to the fact that the lack of a signaling system reduces the probability of detecting these conditions. When such

149

PTCIP

conditions are present in signaled territory, on the other hand, the signal governing movement through the affected block will display a restrictive aspect, thereby providing increased protection in these territories relative to non-signaled TWC territories. Implementation of PTC will reduce the risk associated with misaligned switches in non- signaled TWC territories due to the fact that each switch will be equipped with a wayside interface unit (WIU) to monitor the switch‟s position, with PTC enforcing a positive stop in advance of a misaligned switch. While the lack of track circuits in non-signaled TWC territories also decreases the likelihood of detecting broken rail or fouling conditions, in this risk prioritization exercise it was assumed that, for non-signaled TWC territories, the PTC system will not noticeably reduce the risk associated with broken rail or fouling equipment detection capabilities. Due to the fact that PTC will not offer increased mitigation of these risks, the lack of broken rail and fouling equipment detection was not taken into account when defining the factor levels and estimating the weighting for this risk factor. According to page 11077 of the preamble of 49CFRPart 236 Subpart H, FRA research through the Volpe centre indicated that “risk per train mile in dark territory is approximately 2 times the risk of other territories, TCS, ABS, and Auto,” where “Auto” refers to high-performance signaling systems, such as automatic train stop (ATS) and automatic cab signal (ACS)) [9], and page 2646 of the preamble to Part 236 Subpart I [1], when referring to non-signaled territory, also states that “On a train-mile basis, these operations present about twice the risk as similar operations on signalized lines.” It is assumed that this statement is based on the analysis documented in the Base Case Risk Assessment [6] presentation discussed in Section 2.2.1.2 above. An October 2008 “Fact Sheet” issued by FRA [10] also indicates that dark territory is less safe than other modes of operation, as it describes four “basic types of configuration” and states that “Each type of signaled system has a successively higher level of functionality and redundancy which builds upon the previous with additional layers of safety,” as it refers to dark territory at the bottom of this hierarchy of different types of signaled territories. 2. Automatic Block Signal (ABS): An ABS system is a series of consecutive blocks governed by signals, which are activated by a train or by certain conditions (e.g., switch position) that affect the block use. In the case of CN, blocks in ABS territories are only governed by wayside block signals; not cab signals. As noted in the October 2008 FRA “Fact Sheet”, “The signal displays or indications of an ABS system do not actually provide a locomotive engineer with authority for train movements but rather supplement another form of permission. For example, in a territory equipped with an ABS system, a verbal authority2 is given by the dispatcher for a train to proceed from Point A to Point B, as in non-signaled territory. However, the area is also equipped with wayside signals which provide information regarding: rail integrity (broken or missing rail), train occupancy (a train in the “block” or section of track governed by the signal), and/or the position of track switches in the block.” Signals provide train operators with overt visual information about the conditions of the track ahead, thereby achieving reduced risk relative to subdivisions operated under non-

2 It is assumed that FRA‟s reference to “verbal authority” is actually intended to address “written authorizations that are verbally communicated”.

150

PTCIP

signaled TWC. While signals provide operators with an indication of the condition of the track ahead, it is still possible for operators to disregard signal indications, creating the opportunity for an accident to occur. Implementation of PTC will be expected to provide increased alerting and enforcement of the limits of authority, thereby reducing the risk associated with this method of operation. As previously mentioned, operations in ABS also generally have decreased risk relative to operations in non-signaled TWC territories due to the fact that these signaled territories offer increased probability of detecting misaligned switches, fouling equipment, and broken/missing rails. When these conditions are present in signaled territory, the signal governing movement through the affected block would be expected to display a restrictive aspect, and a train encountering the restrictive signal at the entrance to the block would be required to operate at restricted speed through the entire block. In signaled territory, implementation of PTC will reduce risk relative to current operation by continuously displaying the speed to be maintained and enforcing the upper limit of the railroads‟ restricted speed rule (not to exceed 20 mph). However, the reductions in risk achieved through the implementation of PTC in ABS territory are not expected to be as significant as the reductions achieved in non-signaled TWC territory. 3. Traffic Control System (TCS): Signal indications in TCS territory actually provide the authority for train movement along with the protections described for ABS above. The October 2008 FRA “Fact Sheet” notes that in TCS territory (, “…the signals at each control point are managed directly by a dispatcher and there may be one or more “automatic signals” placed along the stretch of track between the dispatcher-controlled signals at adjacent control points.” In these territories the authority for train movements comes from a dispatcher to a train crew via the indications of the wayside signals that the dispatcher directly controls. The “Fact Sheet” also indicates that TCS technology “…utilizes vital wayside signal system logic to ensure that route integrity is preserved and appropriate timing is provided prior to signals displaying proceed indications in order to keeps trains apart.” This method of operation is generally considered to offer slightly reduced risk in comparison with ABS. The reductions in risk resulting from introduction of PTC on TCS territories are also expected to be similar to the reductions that will be experienced on ABS track, but perhaps slightly reduced.

Implementation of PTC would be expected to achieve a reduction in risk relative to all of the methods of operation identified above, as the enforcement provided by PTC will help compensate for the effects of fatigue and other human factors that contribute to mishaps that are more likely to occur when these standard methods of operation are employed without PTC. As stated in the October 2008 FRA “Fact Sheet”, “The type of wayside signals described above normally function with a very high degree of reliability providing a significant margin of safety. However, if a locomotive engineer is distracted, misperceives or ignores a signal indication, he/she may operate the train above the maximum allowable speed and thus be unable to stop when required.” The added capability of detecting misaligned switches and enforcing stops before misaligned switches will provide substantial reductions in risk in non-signaled TWC territories, while enforcement of speed restrictions in advance of misaligned switches, broken rails, and fouling conditions will provide some benefit in signaled (ABS and TCS) territories.

151

PTCIP

The reduction in risk is expected to be greatest for subdivisions operated under non-signaled TWC, decreased for ABS, and slightly lower for TCS. 3.5.2. Quantification of Risk Factor Weight Taking into consideration the information discussed in the „Overview‟ section above, and considering how this 49CFR risk factor compares with the other risk factors described in Section 3.0, it seemed that this risk factor should receive a moderate-to-high weight. The Volpe research identified inPart 236 Subpart H and discussed in Section 2.2.1.2 above indicates that risk per train mile in dark territory is 2 times the risk of other territories (and this was again noted in Part 236 Subpart I). This seemed to be a key statistic to keep in mind, particularly considering the fact that the analysis that provided this result focused on PTC-preventable accident data. Still, the RSAC Report on Implementation of Positive Train Control Systems [5] discussed in Section 2.2.1.1 concluded that “The train control method was less important in prediction of the accidents of interest in this dataset than other factors.”

Also, starting on page 2618 of the preamble to Part 236 Subpart I [1], the FRA discusses the “Rail Route Analysis Rule” and notes that “No methodology is currently specified for evaluating the potential impact of a PTC system (which would vary in risk reduction depending upon the underlying or previous method of operation). Under these circumstances, there is a distinct possibility the railroads may not give sufficient weight to train control (existing or planned),” and in a footnote to this statement, FRA states that “At least one Class I railroad consolidated some of its PIH traffic on signalized lines prior to adoption of the Rail Route Analysis Rule. This reflects a recognition that method of operations matters…” While these comments are directed towards the Rail Route Analysis Rule, they seem to support the position that the „Method of Operation‟ risk factor should be weighted relatively high.

CN assessed this risk factor to have a “high” effect on the probability component of risk and a “low” effect on the consequence component of risk. The other railroads generally seemed to draw similar assessments. As a result, the total risk weight assigned to this risk factor was 10.5%. As can be seen in Table 2.1, this weight is essentially the same as that assigned to Number of Tracks. 3.5.3. Quantification of Risk Factor Levels The method of operation was identified for each subdivision. Subdivisions were assigned to one of five factor levels according to Table 3-5 below. For cases in which more than one method of operation is employed on a single subdivision, the risk factor level was determined by determining the percentage of each method of operation on the subdivision. A subdivision needed to have more than 50% of its route miles as TCS or ABS in order to achieve a lower risk rating classification.

Factor Level Description Level 1 ATC/ACS Level 2 TCS Level 3 Bi- Directional ABS Level 4 Directional ABS

152

PTCIP

Level 5 TWC Table 3-5. Factor Levels for ‘Method of Operation’

3.6. Risk Factor #6: Speed of Train Operations 3.6.1. Risk Factor Overview This risk factor provides an indication of the maximum authorized speed with which trains are permitted to travel on a subdivision. The probability of an accident would likely increase as the maximum authorized speed increases due to the fact that increased speeds require greater stopping distances, and provide train operators with less time to react to a potentially-hazardous condition. Signal overruns are generally more likely when traveling at higher speeds.

While increased operating speed could potentially act in combination with other risk factors, such as grade or curvature, for example, in causing an accident, it is recognized that the maximum authorized speed is likely to be lower in areas where other such risk factors are increased in order to compensate for the presence of these other risk factors. For example, a segment of extreme curvature would likely already have a relatively low maximum authorized civil speed. The dependency of maximum authorized speeds on track class is another example of where lower speeds are required in areas of higher risk, in this case due to track quality. Similarly, higher maximum authorized speeds generally correspond with the presence of more advanced signaling systems. As a result, when considering this risk factor, it was assumed that some trade-offs exist where the maximum authorized speeds are higher, and that while trains may be permitted to travel at higher speeds on certain segments, it is due to the fact that these segments are inherently safer than segments where lower speeds are required.

PTC systems would be expected to reduce both the accident rate and the consequences associated with this risk factor, as PTC should be able to prevent the occurrence of many would-be accidents, and in cases where PTC is not able to totally prevent an accident from taking place, it should at least limit the consequences by maintaining train speed under prescribed limits. The four primary types of mishaps addressed by Subpart I (overspeed derailments, train collisions, incursions into work zones, movement through misaligned switches) all have greater risk of occurring where trains operate at higher speeds,

153

PTCIP which indicates that PTC will achieve greater risk reduction in higher speed segments than in lower speed segments.

3.6.2. Quantification of Risk Factor Weight Taking into consideration the information discussed in the „Overview‟ section above, and considering how this risk factor compares with the other risk factors described in Section 3.0, it seemed that this risk factor should receive a moderate weight relative to the other “minimum critical” risk factors. The report Risk and Train Control: A Framework for Analysis [7] corroborates the assessment in the „Overview‟ section above, as it notes that “…higher train speeds increase both the likelihood and the severity of the consequences of accidents if there is a signal overrun or a failure to obey a slow order.”

The RSAC Report on Implementation of Positive Train Control Systems [5] provided an analysis that concluded that areas of increased speed are associated with lower accident frequency, stating that “It is counterintuitive to think that accidents decrease with speed limit increases as suggested by the parameter on length-weighted average speed. However, we might reverse the description of this variable and say that we have imposed lower speed limits where accident risk is higher; if we had the luxury of looking at a time-series model we may notice that speed limit changes have taken place over time where risk factors were present.” While the results of the analysis in this report are being considered with caution, this conclusion does not seem unreasonable.

CN assessed this risk factor to have a “medium” effect on the probability component of risk and a “high” effect on the consequence component of risk. The other railroads generally seemed to draw similar assessments. As a result, the total risk weight assigned to this risk factor was 11%. As can be seen in Table 2.1, this weight is essentially the same as that assigned to Number of Tracks and Method of Operation. 3.6.3. Quantification of Risk Factor Levels The maximum authorized speed was identified for each segment of each subdivision and the maximum speed was used. For segments where the maximum authorized speeds vary depending on train type (passenger, freight, etc.), the maximum authorized speed for any train type was used.

Limits for five factor levels provided in Table 3-6 below were assigned taking the range of maximum authorized speed data into account.

Table 3-6. Factor Levels for ‘Speed of Train Operations’ Factor Level Lower Limit Upper Limit Level 1 0 MPH <15 MPH Level 2 15 MPH <30 MPH Level 3 30 MPH <60 MPH Level 4 60 MPH <80 MPH

Level 5 80 MPH None

154

PTCIP

3.7. Risk Factor #7: Grade 3.7.1. Risk Factor Overview This risk factor addresses the slope of the railroad track, as track grade is an expression of the percentage of a track‟s rise for the length of its run. Grade would primarily affect the probability component of risk due to the fact that instances of higher grade may increase train momentum to the point that a train could “get away”, preventing braking from stopping the train, or in less extreme cases, the increase in momentum could at least make it more difficult for a moving train to come to a stop within the limit of authority. Increased grade could also act to indirectly increase the consequences of an accident by increasing the speed and kinetic energy of the train, leading to increased damage upon collision.

PTC would be expected to reduce the risk associated with instances of higher grade, as it would be expected that PTC will enforce speed restrictions, taking the known grade of a segment of track into account in the braking algorithm. This will allow control of the train to be maintained and will allow the train to be stopped within the limit of authority. Reductions in risk, therefore, would be expected to be greater in subdivisions containing segments of higher grade. 3.7.2. Quantification of Risk Factor Weight Taking into consideration the information discussed in the „Overview‟ section above, and considering how this risk factor compares with the other risk factors described in Section 3.0, it seemed clear that this risk factor should receive a low weight compared to the other “minimum critical” risk factors. While the report Risk and Train Control: A Framework for Analysis [7] includes grade in the modeling of the hypothetical corridor, this risk factor was not a focus of the study, was not varied in the sensitivity analyses, and was not addressed in the findings of the study. This risk factor was also not addressed in the other studies documented in Section 2.2.1, and this fact arguably may serve as some indication of its lack of significance on risk.

CN assessed this risk factor to have a “medium” effect on the probability component of risk and a “medium” effect on the consequence component of risk. The other railroads generally seemed to draw similar assessments. As a result, the total risk weight assigned to this risk factor was 4%.

3.7.3. Quantification of Risk Factor Levels For each subdivision, the maximum grade for any portion of the subdivision was used as a measure of grade for the subdivision. Limits for five factor levels, provided in Table 3-7 below, were assigned taking the range of grade data into account.

Table 3-7. Factor Levels for ‘Grade’ Factor Level Lower Limit Upper Limit Level 1 0.0% <0.5% Level 2 0.5% <1.0%

Level 3 1.0% <1.5% Level 4 1.5% <2.0% Level 5 2.0% None

155

PTCIP

3.8. Risk Factor#8: Curvature 3.8.1. Risk Factor Overview This risk factor provides an indication of the degree of arc in the railroad track. Risk of an accident would be expected to increase as curvature increases due to the fact that the centrifugal force to the rail at the point of wheel contact is a function of the severity of the curve, the speed of the vehicle, and the mass of the vehicle. When this force at the point of contact becomes too great, it may overcome the resistance of the track and the train may derail due to rail rollover, the car rolling over, or from the combined transverse force exceeding the limit allowed by rail-flange contact [13]. Instances of increased curvature create an increased potential for an accident because they contribute positively to this increased centrifugal force. It should be noted, however, that super elevation is a common design practice that works to reduce the effects described above. While super elevation may reduce the risk associated with instances of increased curvature, only the degree of curvature was considered in this assessment. This was considered a reasonable approach due to the fact that, as described in the subsection following, this risk factor was not estimated to have a significant effect on risk either way.

PTC would be expected to reduce the risk associated with instances of higher curvature, as it would be expected that PTC will enforce civil speed restrictions, which should account for the degree of track curvature. This will help ensure that trains travel through areas of curvature at a safe operating speed, thereby decreasing the likelihood of derailment. Reductions in risk, therefore, would be expected to be increased in subdivisions containing higher numbers of curves and instances of significant curvature. 3.8.2. Quantification of Risk Factor Weight Taking into consideration the information discussed in the „Overview‟ section above, and considering how this risk factor compares with the other risk factors described in Section 3.0, it seemed clear that this risk factor should receive a low weight compared to the other “minimum critical” risk factors. The RSAC Report on Implementation of Positive Train Control Systems [5] included curvature parameters in the regression analysis, as it assessed both number of curves per mile and the presence/absence of curvature on the territory. The analysis concluded that both of these factors are “not significant”, meaning that they do not have a noticeable effect on risk. The report Risk and Train Control: A Framework for Analysis [7] includes curvature in the modeling of the hypothetical corridor, but this risk factor was not a focus of the study, was not varied in the sensitivity analyses, and was not highlighted in the findings of the study. The study indicated that other risk factors, such as „Presence and Volume of Passenger Traffic‟, „Number of Tracks‟, „Annual MGT Level‟, and „Speed of Train Operations‟ were more significant factors than „Curvature‟.

CN assessed this risk factor to have a “medium” effect on the probability component of risk and a “negligible” effect on the consequence component of risk. The other railroads generally seemed to draw similar assessments. As a result, the total risk weight assigned to this risk factor was 2%.

156

PTCIP

3.8.3. Quantification of Risk Factor Levels For each subdivision, the maximum curvature for any section of the subdivision was used as a measure of curvature on the subdivision. Limits for five factor levels, provided in Table 3-8 below, were assigned taking the range of curvature data into account.

Table 3-8. Factor Levels for ‘Curvature’ Factor Level Lower Limit Upper Limit Level 1 0.0° <2.0° Level 2 2.0° <4.0°

Level 3 4.0° <6.0°

Level 4 6.0° <8.0° Level 5 8.0° None

3.9. Other Risk Factors Not Included in the Risk Prioritization Model This section contains examples of various other risk factors that were considered for use in the risk prioritization model, but which were ultimately not included in the model. Along with each identified risk factor are brief comments on how the factor is estimated to affect risk and on why the factor was not included in CN‟s risk prioritization model.  Rail Traffic Density: This risk factor provides a measure of the number of trains that travel on a segment of track, taking into account the length of the track segment. Due to the fact that this risk factor overlaps considerably with the „Annual MGT Level‟ risk factor („Rail Traffic Density‟ and „Annual MGT Level‟ offer similar information), it was not included in CN‟s risk prioritization model.  Trip Length for Route: This risk factor addresses the lengths of the subdivisions that were assessed in the risk prioritization model. The lengths of the subdivisions would be of concern when assessing the absolute risk associated with subdivisions due to the fact that longer subdivisions would generally be associated with higher absolute risk values (longer track would be associated with higher absolute accident frequencies, and therefore, increased risk of equipment damage, fatalities, injuries, and other economic costs). Assessment of absolute risk was not a primary concern when assessing the priority with which subdivisions should be equipped, as it was not CN‟s intent to discount shorter subdivisions that may have other high-risk attributes. As a result, this risk factor was not included in CN‟s risk prioritization model.  Track Type, Class and Maintenance Schedule: The risk factors identified here are primarily associated with the integrity of the operating environment. While these factors may have an effect on the frequency with which accidents occur, the CN subdivisions to be equipped with PTC do not vary significantly on the basis of these factors, and in comparison with other factors that were included in the risk prioritization model, these factors were not considered to be primary factors of importance for the purposes of assessing the prioritization of PTC implementation.  Presence or Absence, and Types, of Wayside Hazard Detectors: §236.1005(c) requires that all hazard detectors integrated into a signaling system be integrated into the PTC system and that their warnings be appropriately and timely enforced. In the context of this risk factor, subdivisions that lack hazard detectors or that have hazard detectors which are not integrated into the signaling system receive no reduction in risk as a result of installing PTC, while subdivisions with hazard detectors integrated into the signaling system may experience some reduction in risk

157

PTCIP

as the indications provided by the detectors will be enforced by the PTC system. The type of enforcement and the extent to which enforcement reduces risk would vary by the type of hazard detector. CN currently utilizes wayside impact load detectors (WILDs), hot box detectors (HBDTs), and high wide load detectors (HWLDs) on its US network, but these various types of detectors are not integrated into the signaling system. This risk factor, therefore, is not applicable to CN‟s operations and it was not included in the risk prioritization model.  Highway-to-Rail Grade Crossings: 236.1005(a)(4)(ii) requires that the PTC system shall provide an appropriate warning or enforcement when a mandatory directive is issued associated with a highway-rail grade crossing warning system malfunction as required by §§234.105 (warning system activation failure), 234.106 (warning system partial activation), or 234.107 (warning system false activation). When these malfunctions are detected and reported to the train dispatcher, they are protected through the issuance of mandatory directives to the operating crews. Through enforcement of these mandatory directives, a minor reduction in risk related to highway-to-rail grade crossings may be achieved as a result of PTC installation, with greater reductions achieved on subdivisions with greater numbers of highway-to-rail grade crossings relative to subdivisions with smaller numbers of such crossings, this risk factor was considered to be of very low importance relative to other risk factors that were included in CN‟s risk prioritization model, and therefore, this risk factor was not included in the model.  Passenger Stations: The presence of passenger stations increases the risk associated with a subdivision, as stations are typically areas of congregation which may, at times, result in high concentrations of people located relatively close to the track. Accidents occurring at or near passenger stations, particularly accidents involving TIH/PIH or other hazardous materials, could potentially have significantly increased consequences relative to accidents occurring on other portions of the track, away from passenger stations. Instead of including this as a separate risk factor, the risk associated with passenger stations was considered as a part of the „Presence and Volume of Passenger Risk Factor‟, which was previously described in Section 3.2 above.  Venues of Mass Congregation: This risk factor addresses the potential for an accident to occur in an area close to an unusually high concentration of people. „Venues of Mass Congregation‟ include locations such as sports stadiums, concert pavilions, amusement parks, and other locations where unusually high concentrations of people may be found. Again, accidents involving TIH/PIH or other hazardous materials would be the primary accidents of concern when considering these venues. While the consequences associated with accidents that occur near such venues could be significant, this risk factor was considered of very low importance relative to other factors that were included in the risk prioritization model due to the fact that 1) these venues are typically, although not always, found in areas of higher population density and „Population Density‟ is already included in the model, and 2) the probability of a PTC-preventable accident involving a release of hazardous materials occurring near a venue of mass congregation at the exact time that people are congregating in the venue (during a sporting event, concert, etc.) was estimated to be very low.  Presence of Non-PTC-Equipped Traffic Along Shared Route: After December 31, 2015, CN does not intend to have any non-PTC-equipped traffic operating on its subdivisions with PTC deployed, with the only exception being the presence of non-PTC-equipped trains crossing CN track at certain diamond crossings.  Past Accident/Incident Statistics: For the purposes of this risk prioritization model, the most applicable historical data that could be taken into consideration is PTC-preventable accident data for the CN subdivisions being assessed (assuming the data was collected under current operation conditions). Historical data for CN subdivisions were collected and reviewed, but these data, which were few, did not provide significant insight into how installation of PTC would be

158

PTCIP

expected to reduce relative risk on CN subdivisions, and therefore, the data were not included in the risk prioritization model.  Hazards to Human Health and the Environment: According to the Risk Prioritization Template, this risk factor, and specifically the Non-Accident Release Risk Index (NARRI), is intended to be used as a means of evaluating the impact of TIH and other critical hazards materials and their impact on human health and the environment. While NARRI does characterize the nature of non-accident releases, focusing on the severity of the releases, CN did not include this risk factor in its risk prioritization model because the „Presence and Volume of TIH/PIH Materials‟ risk factor sufficiently capture the risk associated with TIH/PIH and other hazardous materials for the purposes of this assessment.

159

PTCIP

4. Model Calculation Tool

Table 4-1. Risk Factor Weighting Risk Factor Effect on Effect on Probability Consequence Total Risk Probability Consequences Score Score Score Factor Weight Annual million gross ton (MGT) levels Very High Medium 1 0.1 1.1 21.0% Presence and volume of passenger traffic Medium Very High 0.1 1 1.1 21.0% Presence and volume of TIH/PIH Low Very High 0.05 1 1.05 20.0% material (loads and residue) transported Number of tracks High Low 0.5 0.05 0.55 10.5% Method of operation High Low 0.5 0.05 0.55 10.5% Speeds of train operations Medium High 0.1 0.5 0.6 11.4% Grade Medium Medium 0.1 0.1 0.2 3.8% Curvature Medium Negligible 0.1 0 0.1 1.9% Total 5.25 100.0%

Notes on Risk Factor Rankings: 1. Risk factors listed are the minimum critical risk factors identified in the FRA template: “Risk Prioritization Methodology for PTC Systems Implementation”, 2. The above listed risk factors are used to perform the baseline prioritization of CN PTC required line segments, 3. Calculation of Final Risk Factor Weightings is based on the following formula: Risk Factor Weight (%) = 100 x ( Probability Score + Consequence Score) / (Total Score of all Risk Factors) 4. Risk factor evaluation of probability and consequences are based on input from subject matter experts from the Operations and Engineering departments of CN.

Risk Factor Weights

Levels Score Very High 1 High 0.5 Medium 0.1 Low 0.05 Very Low 0.01 Negligible 0

160

PTCIP

Table 4-2 Risk Factor Ranges Range Values Risk Factor Unit 0 1 2 3 4 5 MGT per year <5 MGT 5 - <15 MGT 15 - <25 MGT 25 - <35 MGT 35 - <45 MGT 45 MGT & over Passenger Trains trains / day 0 1 - 2 3 - 5 6 - 10 11 - 20 >20 TIH Volumes cars / year 0 1 - <100 100 - <1000 1,000 - <5,000 5,000 - <10,000 GT 10,000 Number of Tracks main tracks N/A GT 2 tracks N/A 2 tracks N/A 1 track Method of Ops N/A ATC / ACS TCS Bi-Dir ABS Dir ABS No Sig/TWC Train Speed max train speed mph N/A 0 - <15 15 - <30 30 - <60 60 - <80 80 & over Track Grade max. grade percent N/A LT 0.5% 0.5% - <1.0% 1.0% - <1.5% 1.5% - 2.0% 2.0% & over Track Curvature max curvature degrees N/A 0 - <2.0 2.0 - <4.0 4.0 - <6.0 6.0 - <8.0 8.0 & over Overall Priority weighted average 0 - <1.5 1.5 - <2.0 2.0 - <2.5 2.5 - <3.0 3.0 - <3.50 3.5 & over

Notes: 1. MGT – Use weighted average of MGT for subdivision when multiple measurement sections available, 2. Passenger Trains – Any passenger traffic on the subdivision establishes the passenger risk ranking for the entire subdivision, 3. TIH Volume – Use maximum number of TIH cars (loads + residue) on subdivision if there are multiple measurement sections. 4. Number of Tracks – Must have more than 50% of route miles as double or multi-track to score lower risk rating for subdivision, 5. Method of Operations – Must have more than 50% of route miles as TCS or ABS to achieve lower risk rating for subdivision, 6. Train Speed – Use maximum train speed on subdivision (passenger, intermodal or freight), 7. Track Grade – Use maximum grade on subdivision (percent), 8. Track Curvature – Use maximum curvature on subdivision (degrees), 9. Overall Priority – Sum of values for individual risk factors for a subdivision.

Table 4-3 Risk Prioritization Model PTC Primary Risk Factor Ratings

161

PTCIP

PTC Primary Risk Factor Ratings

1 2 3 4 5 6 7 8 9 MGT 2008 Passenger TIH 2008 Num Trks Meth. Ops Train Speed Grade Curves Summary Weighted Value Weight Value Weight Value Weight Value Weight Value Weight Value Weight Value Weight Value Weight Priority Subdivision (0-5) 0.210 (0-5) 0.210 (0-5) 0.200 (1-5) 0.105 (1-5) 0.105 (1-5) 0.110 (1-5) 0.040 (1-5) 0.020 Holly 1 0.210 3 0.630 3 0.600 5 0.525 2 0.210 4 0.440 2 0.080 4 0.080 2.775 Joliet 0 0.000 4 0.840 3 0.600 3 0.315 2 0.210 4 0.440 1 0.040 3 0.060 2.505 Leithton 2 0.420 0 0.000 3 0.600 5 0.525 2 0.210 3 0.330 2 0.080 5 0.100 2.265 Matteson 1 0.210 0 0.000 4 0.800 3 0.315 4 0.420 3 0.330 2 0.080 5 0.100 2.255 McComb 2 0.420 1 0.210 5 1.000 5 0.525 2 0.210 4 0.440 1 0.040 2 0.040 2.885 Memphis 2 0.420 1 0.210 5 1.000 5 0.525 5 0.525 3 0.330 1 0.040 3 0.060 3.110 Minneapolis 1 0.210 0 0.000 1 0.200 5 0.525 5 0.525 3 0.330 3 0.120 2 0.040 1.950 Missabe 4 0.840 0 0.000 2 0.400 5 0.525 2 0.210 3 0.330 3 0.120 5 0.100 2.525 Mt. Clemens 2 0.420 0 0.000 4 0.800 5 0.525 5 0.525 3 0.330 2 0.080 1 0.020 2.700 Neenah 5 1.050 0 0.000 3 0.600 5 0.525 2 0.210 4 0.440 3 0.120 3 0.060 3.005 P&I RR 5 1.050 0 0.000 3 0.600 5 0.525 2 0.210 3 0.330 5 0.200 1 0.020 2.935 Peoria 1 0.210 0 0.000 3 0.600 5 0.525 5 0.525 3 0.330 3 0.120 4 0.080 2.390 Rainy 4 0.840 0 0.000 3 0.600 5 0.525 2 0.210 4 0.440 3 0.120 3 0.060 2.795 Shelby 2 0.420 1 0.210 5 1.000 3 0.315 5 0.525 4 0.440 1 0.040 2 0.040 2.990 Shore Line 2 0.420 3 0.630 3 0.600 5 0.525 2 0.210 3 0.330 2 0.080 3 0.060 2.855 South Bend 4 0.840 3 0.630 4 0.800 3 0.315 2 0.210 4 0.440 2 0.080 4 0.080 3.395 Sprague (US) 4 0.840 0 0.000 3 0.600 5 0.525 2 0.210 4 0.440 2 0.080 2 0.040 2.735 St.Louis 1 0.210 0 0.000 1 0.200 5 0.525 2 0.210 4 0.440 2 0.080 3 0.060 1.725 Superior 4 0.840 0 0.000 3 0.600 5 0.525 2 0.210 4 0.440 4 0.160 3 0.060 2.835 Valley 1 0.210 0 0.000 2 0.400 5 0.525 5 0.525 3 0.330 3 0.120 4 0.080 2.190 Waterloo 1 0.210 0 0.000 2 0.400 5 0.525 2 0.210 3 0.330 3 0.120 4 0.080 1.875 Waukesha 5 1.050 5 1.050 3 0.600 5 0.525 2 0.210 4 0.440 3 0.120 4 0.080 4.075 Yazoo 5 1.050 1 0.210 4 0.800 5 0.525 2 0.210 4 0.440 2 0.080 3 0.060 3.375 Risk Factor Weights Risk Factor Groupings - Based on Total of Risk Factors 21.0% Annual MGT level 10.5% Method of operation 0 Total Priority <1.5 3 Total Priority 2.5 to <3.0 21.0% Passenger Volume 11.0% Speeds of train operations 1 Total Priority 1.5 to <2.0 4 Total Priority 3.0 to <3.5 20.0% TIH?PIH Volume 4.0% Grade 2 Total Priority 2.0 to <2.5 5 Total Priority 3.5 & over 10.5% Number of Tracks 2.0% Curvature

162

PTCIP

Table 4-4: Risk Factor: Annual Million Gross Ton (MGT) Num Subdivision Total Tonnage MGT Risk Factor 2008 2009 2008 2 Yr Avg 1 Baton Rouge 11,536,101 9,636,942 11.54 10.6 1 2 Beaumont 11,217,275 10,139,536 11.22 10.7 1 3 Bluford 26,061,357 17,692,249 26.06 21.9 3 4 Cairo 22,381,552 18,589,667 22.38 20.5 2 5 Centralia 29,365,359 24,369,179 29.37 26.9 3 6 Champaign 39,584,325 31,526,676 39.58 35.6 4 7 Cherokee 5,655,797 5,184,857 5.66 5.4 1 8 Chicago 39,737,099 30,545,395 39.74 35.1 4 9 Dubuque 11,647,637 10,964,124 11.65 11.3 1 10 Elsdon 25,967,671 25,967,671 25.97 26.0 3 11 Flat Rock 7,797,156 5,765,967 7.80 6.8 1 12 Flint 42,404,327 34,875,491 42.40 38.6 4 13 Fox River 9,149,484 6,777,377 9.15 8.0 1 14 Freeport 9,923,760 10,166,013 9.92 10.0 1 15 Fulton 40,138,303 29,550,689 40.14 34.8 4 16 Gilman * 5,474,443 4,051,088 5.47 4.8 1 17 Hammond 5,449,009 4,870,309 5.45 5.2 1 18 Holly 7,043,036 4,701,505 7.04 5.9 1 19 Joliet 4,263,025 3,532,438 4.26 3.9 0 20 Leithton 19,340,983 19,340,983 19.34 19.3 2 21 Matteson 11,502,691 11,502,691 11.50 11.5 1 22 McComb 24,248,060 18,480,016 24.25 21.4 2 23 Memphis 15,175,524 13,118,881 15.18 14.1 2 24 Minneapolis 5,667,904 3,834,475 5.67 4.8 1 25 Missabe 36,279,071 31,879,272 36.28 34.1 4 26 Mount Clemens 17,257,028 13,759,433 17.26 15.5 2 27 Neenah 52,886,288 44,329,037 52.89 48.6 5 28 P&I RR * 46,586,284 30,226,732 46.59 38.4 5 29 Peoria 5,094,357 4,622,483 5.09 4.9 1 30 Rainy 38,501,834 30,118,484 38.50 34.3 4 31 Shelby 20,958,666 16,499,903 20.96 18.7 2 32 Shore Line 16,221,968 13,814,636 16.22 15.0 2 33 South Bend 42,348,819 36,429,433 42.35 39.4 4 34 Sprague (US) 44,741,265 37,519,311 44.74 41.1 4 35 St.Louis 8,251,079 7,333,575 8.25 7.8 1 36 Superior 39,795,070 33,377,331 39.80 36.6 4 37 Valley * 5,502,262 7,493,348 5.50 6.5 1 38 Waterloo 9,657,393 8,949,040 9.66 9.3 1 39 Waukesha 52,267,775 42,839,513 52.27 47.6 5 40 Yazoo 45,189,959 36,213,542 45.19 40.7 5 Notes

1) Subdivision GTM values are weighted averages for subdivisions with multiple measurement sections 2) * Subdivisions with weighted averages below 5MGT and peak sections over 5MGT use the peak section values 3) EJ&E Subs do not have MGT for 2008 use 2009 data (EJ&E) July-Dec prorated for 12 months. 4) P&I RR uses Bluford Sub tonnages between Chiles Jct & Metropolis Jct. 5) Shelby sub based on average of Yazoo North and Fulton South measurement sections

Risk Ranking Ranges 0 Less than 5 MGT 3 25 - <35 MGT 1 5 - <15 MGT 4 35 - <45 MGT 2 15 - <25 MGT 5 45 MGT and over

163

PTCIP

Table 4-5: Risk Factor: Presence and Volume of Passenger Traffic Num Subdivision Daily Amtrak Daily Commuter Total Passenger Risk Factor Trains Trains Trains

1 Baton Rouge 0 0 0 0 2 Beaumont 0 0 0 0 3 Bluford 0 0 0 0 4 Cairo 2 0 2 1 5 Centralia 6 0 6 3 6 Champaign 6 0 6 3 7 Cherokee 0 0 0 0 8 Chicago 6 0 6 3 9 Dubuque 0 0 0 0 10 Elsdon 2 0 2 1 11 Flat Rock 0 0 0 0 12 Flint 2 0 2 1 13 Fox River 0 0 0 0 14 Freeport 10 6 16 4 15 Fulton 2 0 2 1 16 Gilman 0 0 0 0 17 Hammond 0 0 0 0 18 Holly 6 0 6 3 19 Joliet 10 6 16 4 20 Leithton 0 0 0 0 21 Matteson 0 0 0 0 22 McComb 2 0 2 1 23 Memphis 2 0 2 1 24 Minneapolis 0 0 0 0 25 Missabe 0 0 0 0 26 Mount Clemens 0 0 0 0 27 Neenah 0 0 0 0 28 P&I RR 0 0 0 0 29 Peoria 0 0 0 0 30 Rainy 0 0 0 0 31 Shelby 2 0 2 1 32 Shore Line 6 0 6 3 33 South Bend 8 0 8 3 34 Sprague (US) 0 0 0 0 35 St.Louis 0 0 0 0 36 Superior 0 0 0 0 37 Valley 0 0 0 0 38 Waterloo 0 0 0 0 39 Waukesha 0 22 22 5 40 Yazoo 2 0 2 1

Notes 1) Passenger train volumes use the highest number of passenger trains on any segment of the subdivision.

Risk Ranking Ranges 0 0 passenger trains / day 3 6 - 10 passenger trains / day 1 1 - 2 passenger trains / day 4 11 - 20 passenger trains / day 2 3 - 5 passenger trains / day 5 20 or more passenger trains / day

Table 4-6: Risk Factor: Presence and Volume of TIH/PIH Material (Loads and Residue) Transported Num Subdivision 2008 TIH/PIH Car Counts Other Hazmat Risk Loads Residue Total Loads & Res. Factor 1 Baton Rouge 11,159 11,656 22,815 112,608 5

164

PTCIP

2 Beaumont 4,309 4,256 8,565 27,146 4 3 Bluford 5,549 5,486 11,035 114,279 5 4 Cairo 1,949 2,190 4,139 33,595 3 5 Centralia 2,021 2,267 4,288 34,816 3 6 Champaign 4,976 4,368 9,344 106,644 4 7 Cherokee 18 11 29 12,430 1 8 Chicago 4,112 4,754 8,866 156,239 4 9 Dubuque 320 329 649 39,132 2 10 Elsdon 3,194 3,461 6,655 136,562 4 11 Flat Rock 787 660 1,447 17,318 3 12 Flint 4,138 4,371 8,509 151,187 4 13 Fox River 12 15 27 11,185 1 14 Freeport 1,168 829 1,997 92,319 3 15 Fulton 5,285 5,561 10,846 83,910 5 16 Gilman 27 562 589 387 2 17 Hammond 1,472 3,052 4,524 15,534 3 18 Holly 1,706 1,693 3,399 46,716 3 19 Joliet 2,008 2,207 4,215 97,290 3 20 Leithton 1,123 1,306 1,864 72,623 3 21 Matteson 2,830 3,141 5,233 108,411 4 22 McComb 8,857 8,995 17,852 128,252 5 23 Memphis 5,409 5,896 11,305 108,284 5 24 Minneapolis 10 14 24 6,793 1 25 Missabe 199 530 729 35,988 2 26 Mount Clemens 3,251 3,502 6,753 135,908 4 27 Neenah 1,208 1,382 2,590 81,033 3 28 P&I RR 1,320 1,369 2,689 14,658 3 29 Peoria 1,172 1,075 2,247 6,095 3 30 Rainy 642 809 1,451 84,376 3 31 Shelby 5,285 5,561 10,846 83,910 5 32 Shore Line 1,500 1,464 2,964 43,412 3 33 South Bend 2,854 3,162 6,016 108,708 4 34 Sprague (US) 675 856 1,531 86,352 3 35 St.Louis 16 13 29 2,233 1 36 Superior 1,053 1,224 2,277 96,883 3 37 Valley 407 456 863 8,735 2 38 Waterloo 183 280 463 31,610 2 39 Waukesha 1,174 1,334 2,508 74,567 3 40 Yazoo 4,521 4,578 9,099 98,003 4

Notes: 1) Shelby Sub uses Fulton sub TIH traffic values - no unique data available for Shelby. 2) P&I Railroad uses Bluford sub TIH traffic values - no unique data available for P&I. 3) Matteson and Leithton sub use 2009 TIH traffic values - no data for 2008

TIH Risk Ranking Ranges 0 0 (cars/year) 3 1,000 - <5,000 (cars/year) 1 1 - <100 (cars/year) 4 5,000 - <10,000 (cars/year) 2 100 - <1,000 (cars/year) 5 10,000 or more (cars/year)

Table 4-7: Risk Factor: Number of Tracks Num Subdivision Miles of Main Track Route Track Risk Main 1 Main 2 Main 3 Main 4 Miles Miles Factor 1 Baton Rouge 79.4 79.4 79.4 5 2 Beaumont 168.8 168.8 168.8 5 3 Bluford 163.6 163.6 163.6 5 4 Cairo 41.7 2.2 41.7 43.9 5

165

PTCIP

5 Centralia 115.9 26.8 115.9 142.7 5 6 Champaign 123.1 8.1 123.1 131.2 5 7 Cherokee 127.3 127.3 127.3 5 8 Chicago 122.6 36.3 8.1 5.6 122.6 172.6 5 9 Dubuque 156.9 3.1 156.9 160.0 5 10 Elsdon 16.8 16.8 16.8 33.6 3 11 Flat Rock 23.9 23.9 23.9 5 12 Flint 155.6 68.7 3.1 155.6 227.4 5 13 Fox River 34.9 34.9 34.9 5 14 Freeport 112.2 11.8 112.2 124.0 5 15 Fulton 118.7 17.5 118.7 136.2 5 16 Gilman 28.9 28.9 28.9 5 17 Hammond 42.7 42.7 42.7 5 18 Holly 62.9 26.2 62.9 89.1 5 19 Joliet 33.2 33.2 33.2 66.4 3 20 Leithton 65.0 11.8 65.0 76.8 5 21 Matteson 43.4 23.7 43.4 67.1 3 22 McComb 181.4 36.9 181.4 218.3 5 23 Memphis 2.5 2.5 2.5 5 24 Minneapolis 123.6 123.6 123.6 5 25 Missabe 63.5 7.0 63.5 70.5 5 26 Mount Clemens 45.6 45.6 45.6 5 27 Neenah 86.6 8.6 86.6 95.2 5 28 P&I RR 13.0 13.0 13.0 5 29 Peoria 40.1 40.1 40.1 5 30 Rainy 152.2 152.2 152.2 5 31 Shelby 16.6 16.6 16.6 33.2 3 32 Shore Line 51.0 2.9 51.0 53.9 5 33 South Bend 142.5 117.7 142.5 260.2 3 34 Sprague (US) 43.4 43.4 43.4 5 35 St.Louis 37.1 37.1 37.1 5 36 Superior 233.3 233.3 233.3 5 37 Valley 12.3 12.3 12.3 5 38 Waterloo 103.8 103.8 103.8 5 39 Waukesha 141.7 37.7 5.5 141.7 184.9 5 40 Yazoo 204.1 1.4 204.1 205.5 5

Notes 1) Need to have more than 50% of route miles as double or multi-track to achieve lower risk rating for sub. 2) Track mileages exclude restricted speed tracks segments.

Risk Ranking Ranges 0 Not applicable 1 More than 2 tracks for more than 50% of the route miles on subdivision 2 Not applicable 3 Double track for more than 50% of the route miles on subdivision 4 Not applicable 5 Single track for more than 50% of the route miles on the subdivision

166

PTCIP

Table 4-8: Risk Factor: Method of Operation Num Subdivision Track Miles by Control Method Route Track Risk CTC TWC-ABS TWC-Dark YL/520 Total Miles Miles Factor 1 Baton Rouge 0.7 76.8 - 1.9 79.4 79.4 79.4 4 2 Beaumont - - 168.8 - 168.8 168.8 168.8 5 3 Bluford 121.7 41.9 - - 163.6 163.6 163.6 2 4 Cairo 39.5 4.4 - - 43.9 41.7 43.9 2 5 Centralia 142.7 - - - 142.7 115.9 142.7 2 6 Champaign 131.2 - - - 131.2 123.1 131.2 2 7 Cherokee - - 127.3 - 127.3 127.3 127.3 5 8 Chicago 172.6 - - - 172.6 122.6 172.6 2 9 Dubuque 133.7 - - - 133.7 133.7 133.7 2 10 Elsdon 33.2 0.4 - - 33.6 16.8 33.6 2 11 Flat Rock 23.9 - - - 23.9 23.9 23.9 2 12 Flint 227.4 - - - 227.4 155.6 227.4 2 13 Fox River 6.3 - 28.6 - 34.9 36.3 34.9 5 14 Freeport 20.4 103.6 - - 124.0 112.2 124.0 4 15 Fulton 133.0 - - 3.2 136.2 118.7 136.2 2 16 Gilman - 28.9 - - 28.9 28.9 28.9 4 17 Hammond - - 42.7 - 42.7 42.7 42.7 5 18 Holly 75.1 14.0 - - 89.1 62.9 89.1 2 19 Joliet 66.4 - - - 66.4 33.2 66.4 2 20 Leithton 70.4 6.4 - - 76.8 65.0 76.8 2 21 Matteson 18.4 48.7 - - 67.1 43.4 67.1 4 22 McComb 209.2 3.7 - 5.4 218.3 181.4 218.3 2 23 Memphis - - - 2.5 2.5 2.5 2.5 5 24 Minneapolis - - 123.6 - 123.6 123.6 123.6 5 25 Missabe 36.2 34.3 - - 70.5 63.5 70.5 2 26 Mount Clemens 2.1 - 43.5 - 45.6 45.6 45.6 5 27 Neenah 95.2 - - - 95.2 86.6 95.2 2 28 P&I RR 13.0 - - 1.0 14.0 14.0 14.0 2 29 Peoria - 2.0 38.1 40.1 110.0 40.1 5 30 Rainy 92.1 - 60.1 - 152.2 152.2 152.2 2 31 Shelby 6.7 16.1 - 10.4 33.2 16.6 33.2 5 32 Shore Line 46.9 3.6 - 3.4 53.9 51.0 53.9 2 33 South Bend 260.2 - - - 260.2 142.5 260.2 2 34 Sprague (US) 43.4 - - - 43.4 43.4 43.4 2 35 St.Louis 37.1 - - - 37.1 37.1 37.1 2 36 Superior 233.3 - - - 233.3 233.3 233.3 2 37 Valley 1.0 - 11.3 - 12.3 138.0 12.3 5 38 Waterloo 103.8 - - - 103.8 109.2 103.8 2 39 Waukesha 184.9 - - - 184.9 141.7 184.9 2 40 Yazoo 202.7 - - 2.8 205.5 204.1 205.5 2

Note 1) Tracks must have greater than 50% of track miles using TCS or ABS to be counted as using this method of s operations. 2) Data not available on directional vs bi-directional ABS. Ranking assumes all ABS is directional. 3) Excludes restricted speed track.

Risk Factor Ranking 0 N/A 3 Bi-Directional ABS 1 ATC or ATS 4 Directional ABS 2 TCS 5 TWC

167

PTCIP

Table 4-9: Risk Factor: Speed of Train Operations Num Subdivision Max Train Speed (MPH) Risk Factor Train Passenger Intermodal Freight Maximum Speed 1 Baton Rouge 40 40 3 2 Beaumont 49 49 3 3 Bluford 60 60 4 4 Cairo 79 60 79 4 5 Centralia 79 60 79 4 6 Champaign 79 60 79 4 7 Cherokee 40 40 3 8 Chicago 79 60 79 4 9 Dubuque 50 50 3 10 Elsdon 60 60 60 4 11 Flat Rock 55 55 3 12 Flint 79 60 79 4 13 Fox River 49 49 3 14 Freeport 50 50 3 15 Fulton 79 60 79 4 16 Gilman 60 60 4 17 Hammond 49 49 3 18 Holly 70 60 70 4 19 Joliet 79 60 40 79 4 20 Leithton 45 45 3 21 Matteson 45 45 3 22 McComb 79 60 79 4 23 Memphis 30 25 30 3 24 Minneapolis 40 40 3 25 Missabe 49 49 3 26 Mount Clemens 49 49 3 27 Neenah 60 60 4 28 P&I RR 40 40 3 29 Peoria 40 40 3 30 Rainy 60 60 4 31 Shelby 60 60 4 32 Shore Line 40 40 40 3 33 South Bend 60 60 4 34 Sprague (US) 60 60 4 35 St.Louis 60 60 4 36 Superior 60 60 4 37 Valley 40 40 3 38 Waterloo 50 50 3 39 Waukesha 60 60 60 60 4 40 Yazoo 79 60 79 4

Notes 1) Use maximum authorized train speed on subdivision for risk factor (passenger or freight)

Risk Ranking Ranges 0 N/A 3 30 - <60 (MPH) 1 0 - <15 (MPH) 4 60 - <80 (MPH) 2 14 - <30 (MPH) 5 80 and over (MPH))

168

PTCIP

Table 4-10: Risk Factor: Track Grades Num Subdivision Ruling Grade Grade (%) Milepoint Ascending Direction Risk Factor 1 Baton Rouge 0.50 432.0 North/East 2 2 Beaumont 1.00 149.0 South/West 3 3 Bluford 0.30 102.0 North/East 1 4 Cairo 0.74 394.0 North/East 2 5 Centralia 0.84 335.0 North/East 2 6 Champaign 0.50 175.0 South/West 2 7 Cherokee 1.20 424.0 South/West 3 8 Chicago 0.55 112.0 North/East 2 9 Dubuque 1.40 267.0 North/East 3 10 Elsdon 0.43 9.0 North/East 1 11 Flat Rock 0.30 37.0 South/West 1 12 Flint 0.60 203.0 South/West 2 13 Fox River 0.65 212.0 North/East 2 14 Freeport 0.95 6.5 South/West 2 15 Fulton 0.51 373.0 North/East 2 16 Gilman 1.00 95.0 South/West 3 17 Hammond 0.50 14.0 North/East 2 18 Holly 0.65 21.0 North/East 2 19 Joliet 0.34 23.0 South/West 1 20 Leithton 0.70 45.5 South/West 2 21 Matteson 0.87 8.3 North/East 2 22 McComb 0.45 806.0 North/East 1 23 Memphis 0.40 394.3 South/West 1 24 Minneapolis 1.08 335.5 North/East 3 25 Missabe 1.08 13.1 North/East 3 26 Mount Clemens 0.59 37.0 North/East 2 27 Neenah 1.03 194.0 South/West 3 28 P&I RR 2.00 N/A No Data - Assume 2% 5 29 Peoria 1.12 81.0 North/East 3 30 Rainy 1.20 75.0 North/East 3 31 Shelby 0.49 395.0 North/East 1 32 Shore Line 0.75 47.5 North/East 2 33 South Bend 0.88 55.0 North/East 2 34 Sprague (US) 0.50 30.0 North/East 2 35 St.Louis 0.95 13.0 North/East 2 36 Superior 1.90 472.2 South/West 4 37 Valley 1.04 134.0 South/West 3 38 Waterloo 1.00 373.0 North/East 3 39 Waukesha 1.08 84.0 South/West 3 40 Yazoo 0.50 191.0 North/East 2

Notes on Grade Factors 1) If maximum grade is the same in both directions - show direction with longest grade 3) Subs with no grade data available use 2% grade for maximum risk until data determined.

Risk Ranking Ranges 0 N/A 3 1.0% - <1.5% 1 <0.5% 4 1.50% - <2.0% 2 0.5% - <1.0% 5 2.0% and over

Table 4-11: Risk Factor: Track Curvatures Num Subdivision Total Max Curve Risk

169

PTCIP

Curves (Degrees) Factor 1 Baton Rouge 51 7.63 4 2 Beaumont 224 9.95 5 3 Bluford 78 2.56 2 4 Cairo 42 4.74 3 5 Centralia 113 5.07 3 6 Champaign 52 2.49 2 7 Cherokee 54 3.30 2 8 Chicago 55 4.11 3 9 Dubuque 155 11.93 5 10 Elsdon 23 3.14 2 11 Flat Rock 27 6.82 4 12 Flint 75 4.00 3 13 Fox River 33 5.72 3 14 Freeport 77 4.86 3 15 Fulton 164 4.11 3 16 Gilman 9 2.32 2 17 Hammond 29 11.49 5 18 Holly 45 7.33 4 19 Joliet 12 4.04 3 20 Leithton 10 10.00 5 21 Matteson 53 8.00 5 22 McComb 165 3.91 2 23 Memphis 6 4.04 3 24 Minneapolis 43 3.82 2 25 Missabe 67 8.30 5 26 Mount Clemens 9 1.67 1 27 Neenah 85 5.89 3 28 P&I RR 6 1.96 1 29 Peoria 30 6.18 4 30 Rainy 123 5.00 3 31 Shelby 21 2.18 2 32 Shore Line 49 4.67 3 33 South Bend 73 7.58 4 34 Sprague (US) 8 2.19 2 35 St.Louis 20 5.40 3 36 Superior 152 5.51 3 37 Valley 10 6.25 4 38 Waterloo 70 6.46 4 39 Waukesha 196 7.23 4 40 Yazoo 179 5.05 3

Risk Ranking Ranges 0 N/A 3 4 - <6.0 Degrees 1 0 - <2.0 Degrees 4 6 - <8.0 Degrees 2 2 - <4.0 Degrees 5 8 Degrees & over

170

PTCIP

5. Risk Prioritization Model Results Num Subdivision Avg Peak TIH/PIH Passenger Mainline Risk Weighted MGT MGT Cars Trains/Day Route Miles Factor Priority Group 1 Waukesha 52.3 61.6 2,508 22 141.7 5 4.075 2 Chicago 39.7 43.8 8,866 6 121.9 5 3.585 3 Champaign 39.6 51.5 9,344 6 123.1 5 3.565 4 South Bend 42.3 51.9 6,016 8 142.5 4 3.395 5 Yazoo 45.2 48.4 9,099 2 204.1 4 3.375 6 Fulton 40.1 50.2 10,846 2 118.7 4 3.365 7 Centralia 29.4 39.5 4,288 6 115.9 4 3.175 8 Flint 42.4 53.4 8,509 2 155.6 4 3.165 9 Memphis 15.2 49.8 11,305 2 2.5 4 3.110 10 Freeport 9.9 11.8 1,997 16 112.2 4 3.065 11 Neenah 52.9 57.1 2,590 0 86.6 4 3.005 12 Shelby 20.9 24.0 10,846 2 16.6 3 2.990 13 P&I RR 46.6 46.6 2,689 0 13.0 3 2.935 14 Bluford 26.1 46.6 11,035 0 163.6 3 2.885 15 McComb 24.2 31.2 17,852 2 181.4 3 2.885 16 Shore Line 16.2 41.9 2,964 6 51.0 3 2.855 17 Superior 39.8 55.2 2,277 0 233.3 3 2.835 18 Rainy 38.5 44.5 1,451 0 152.2 3 2.795 19 Holly 7.0 9.0 3,399 6 62.9 3 2.775 20 Sprague (US) 44.7 46.0 1,531 0 43.4 3 2.735 21 Mount Clemens 17.3 17.3 6,753 0 45.6 3 2.700 22 Elsdon 26.0 26.0 6,655 2 16.8 3 2.685 23 Baton Rouge 11.5 15.0 22,815 0 79.4 3 2.645 24 Beaumont 11.2 15.6 8,565 0 168.8 3 2.610 25 Cairo 22.4 22.5 4,139 2 41.7 3 2.545 26 Missabe 36.3 74.5 729 0 63.5 3 2.525 27 Joliet 4.3 6.8 4,215 16 33.2 3 2.505 28 Peoria 5.1 8.0 2,247 0 40.1 2 2.390 29 Hammond 5.4 5.4 4,524 0 42.7 2 2.370 30 Leithton 19.3 43.5 2,429 0 65.0 2 2.265 31 Matteson 11.5 44.4 5,971 0 43.4 2 2.255 32 Valley 3.2 5.5 863 0 12.3 2 2.190 33 Gilman 2.4 5.5 589 0 28.9 2 2.155 34 Flat Rock 7.8 13.2 1,447 0 23.9 1 1.995 35 Cherokee 5.7 6.3 29 0 127.3 1 1.950 36 Minneapolis 5.7 5.7 24 0 123.6 1 1.950 37 Fox River 9.1 20.0 27 0 34.9 1 1.930 38 Dubuque 11.6 14.1 649 0 156.9 1 1.895 39 Waterloo 9.7 9.7 463 0 103.8 1 1.875 40 St.Louis 8.3 12.9 29 0 37.1 1 1.725

Risk Factor Groupings - Based on Weighted Average of Risk Factors 0 Weighted prioirty <1.5 3 Weighted priority from 2.5 to <3.0 1 Weighted priority 1.5 to <2.0 4 Weighted priority from 3.0 to <3.5 2 Weighted priority 2.0 to <2.5 5 Weighted priority 3.5 & over

171

PTCIP

6. External References [1] Federal Railroad Administration, US Department of Transportation. 49CFRParts 229, 234, 235 et al. Positive Train Control Systems; Final Rule. Docket No. FRA-2008-0132, Notice No. 3. January 15, 2010.

[2] Federal Railroad Administration, US Department of Transportation. Risk Prioritization Methodology for PTC System Implementation, January 7, 2010.

[3] Federal Railroad Administration, US Department of Transportation. PTC Implementation Plan (PTCIP) Template, January 7, 2010.

[4] USNRC (United States Nuclear Regulatory Commission) (1990). “Severe Accident Risks: An Assessment for Five U.S. Nuclear Power Plants Final Summary Report,” NUREG-1150, (3 vols).

[5] Railroad Safety Advisory Committee (RSAC), Federal Railroad Administration, US Department of Transportation. Report of the Railroad Safety Advisory Committee to the Federal Railroad Administrator: Implementation of Positive Train Control Systems. September 1999.

[6] Railroad Systems Division (DTS-75), Office of Safety and Security, The John Volpe National Transportation Systems Center, US Department of Transportation. Presentation for Office of Safety, Federal Railroad Administration RSAC/PTC Working Group Risk 2 Team, Base Case Risk Assessment: Data Analysis & Tests. April 22, 2003.

[7] Carl D. Martland, Ying Zhu, Youssef Lahrech, and Joseph M. Sussman. Risk and Train Control: A Framework for Analysis. Center for Transportation Studies, Massachusetts Institute of Technology, Cambridge, January 2001.

[8] Congress of the United States. Rail Safety Improvement Act of 2008. Public Law 110–432. October 16, 2008.

[9] Federal Railroad Administration, US Department of Transportation. 49CFRParts 209, 234, and 236, Standards for Development and Use of Processor-Based Signal and Train Control Systems; Final Rule. Docket No. FRA-2001-10160. March 7, 2005.

[10] Federal Railroad Administration, US Department of Transportation. Signals and Train Control Fact Sheet. October 2008.

[11] Federal Transit Administration, US Department of Transportation. Commuter Rail Safety Study. November 2006.

[12] Richard F. Healing, Member of National Transportation Safety Board, and Chairman of the Board of Inquiry. Opening Statement in Public Hearing on Collision of Union Pacific Railroad Freight Train MHOTU-23 and BNSF Railway Company Freight Train MEAPTUL-126D, Macdona, Texas. April 26-27, 2005. http://www.ntsb.gov/events/2005/Macdona/opening.htm

[13] American Railway Engineering and Maintenance-of-Way Association, Practical Guide to Railway Engineering. AREMA 2003.

172

PTCIP

[14] Saat R. and Barkan C., University of Illinois. Positive Train Control Toxic Inhalation Hazard Risk Analysis Methodology Presentation. January 27, 2010.

[15] Federal Railroad Administration, Office of Safety Analysis. Accident Data (1998 through 2008). http://safetydata.fra.dot.gov/officeofsafety/. Data downloaded on February 21, 2010.

[16] Transport Canada, U.S. Department of Transportation, Secretariat of Transport and Communications of Mexico (SCT) et al. 2008 Emergency Response Guidebook.

173

PTCIP

Appendix C: Review of Previous Applicable Studies

174

PTCIP

Introduction This appendix briefly summarizes reviews of previous research related to the railroad risk factors that are being considered in the risk prioritization model. The three key references that are discussed in this appendix are identified in the bulleted list below (note that the reference numbers included throughout this appendix refer to references listed in Section 6.0 above).  Railroad Safety Advisory Committee (RSAC), Federal Railroad Administration, US Department of Transportation. Report of the Railroad Safety Advisory Committee to the Federal Railroad Administrator: Implementation of Positive Train Control Systems. September 1999. [5]  Railroad Systems Division (DTS-75), Office of Safety and Security, The John Volpe National Transportation Systems Center, US Department of Transportation. Presentation for Office of Safety, Federal Railroad Administration RSAC/PTC Working Group Risk 2 Team, Base Case Risk Assessment: Data Analysis & Tests. April 22, 2003. [6]  Carl D. Martland, Ying Zhu, Youssef Lahrech, and Joseph M. Sussman. Risk and Train Control: A Framework for Analysis. Center for Transportation Studies, Massachusetts Institute of Technology, Cambridge, January 2001. [7]

For each of the above-cited references, the sections following provide a brief overview of the contents of the reference, a listing of the risk factors addressed by the reference, a listing of relevant results/findings cited in the reference, and a discussion of how results/findings were considered when assigning risk factor weights for use in the risk prioritization model.

Report of the Railroad Safety Advisory Committee: Implementation of Positive Train Control Systems Overview: The RSAC Report of the Railroad Safety Advisory Committee to the Federal Railroad Administrator: Implementation of Positive Train Control Systems [5] made use of historical data to help estimate PTC benefits by evaluating PTC-preventable accidents (PPAs). RSAC identified 819 PPAs out of the more than 25,000 accidents reported to FRA between 1988 and 1997 and attempted to estimate the benefits of PTC by evaluating how many accidents would have been eliminated were a given type of PTC system in place. These data were input into the Corridor Risk Assessment Model (CRAM) model, which was used to estimate the safety benefits of PTC by relating the historic occurrence and consequences of accidents that may have been prevented by a PTC system to specific track features and traffic.

Risk Factors Addressed: The independent variables included in the regression analyses from the RSAC report are listed below: 1) The log of the total number of trains on a segment per year (the sum of passenger and freight trains) 2) The squared log of the total number of trains on a segment per year 3) The presence/absence of multi-track (variable set equal to 0 for single track and 1 for multi-track) 4) The ratio of passenger trains to total trains 5) The number of switches per mile 6) The number of curves per mile 7) The presence/absence of curvature on the territory 8) The length-weighted average curvature for the territory

175

PTCIP

9) The percentage of territory miles under automatic train stop (ATS) or automatic cab signaling (ACS) control 10) The percentage of territory miles signaled, but not under ATS or ACS control 11) The length-weighted average speed for a territory.

Key Findings/Results: The bulleted list below identifies findings of this study that appear to be relevant to the risk factors considered in the risk prioritization model.  Page 79 of the RSAC report states that “…the analysis pointed out that, of the corridors studied, the highest predictor of risk was the volume of traffic (as expressed by the log squared of the total trains per year). The train control method was less important in prediction of the accidents of interest in this dataset than other factors.”  The table presenting results of the analysis of which factors are most influential in determining the occurrence of PTC-preventable accidents (on page 75) indicates that factors that were “not significant” (at confidence level of 0.05) in determining the occurrence of PTC-preventable accidents for all four levels of PTC technology were the log of the total number of trains on a segment per year, the ratio of passenger trains to total trains, curves per mile, and presence/absence of curvature.  Pg. 78 of the report states that “Using the highest level of PTC, the model indicates that the total train flow, the number of tracks, and the number of switches and curves per mile contribute to increases in the expected number of accidents and that the presence of a train control method higher than dark but lower than automatic train control will reduce that risk. In addition, two other factors contribute to lowered risk, the average length of curves at a location and the average maximum allowable speed.”  Page 79 also states that “It is counterintuitive to think that accidents decrease with speed limit increases as suggested by the parameter on length-weighted average speed. However, we might reverse the description of this variable and say that we have imposed lower speed limits where accident risk is higher; if we had the luxury of looking at a time-series model we may notice that speed limit changes have taken place over time where risk factors were present.”  One additional statement included in this report (on page 69) that is worth noting is one that relates to the risk of hazardous materials: “The trends in the derailment category indicate relatively infrequent low-consequences events, whose greatest potential hazard is in the possibility of the release of hazardous chemicals requiring an evacuation. Seventeen of four hundred twenty derailments resulted in evacuations; the average number of people evacuated was approximately 420 per incident. Two incidents resulted in over 1000 evacuations. One derailment, included in the group of accidents thought to be possibly preventable by the highest level of PTC system, accounted for 47 fatalities. This accident is not consistent with the general trend of the consequences of PTC-preventable derailments being less than collisions, but it identifies a source of risk.”

General Comments on How Results Were Used In Estimation of Risk Factor Weights: Some of the details of the regression analyses that led to the findings/results identified in the bulleted list above are not included in the RSAC report, and therefore, the findings/results were considered with some caution. There is nothing, however, to indicate that the analysis contained in this report is not valid or useful.

176

PTCIP

In general, the conclusions of the report indicate that the „Annual MGT Level‟ risk factor should potentially be assigned a relatively high weight. The „Curvature‟ risk factor may not be a very strong indicator of risk, so it should potentially be assigned a relatively low weight. „Speed of Train Operations‟ actually had a counter-intuitive effect on predicting the occurrence of accidents, but this may not be unreasonable due to the fact that authorized speeds may be lower/higher on a section of track dependent on the presence/absence of other risk factors. The data presented in the final bullet help illustrate the noticeable effect on risk that „Presence and Volume of TIH/PIH Material‟ and „Presence and Volume of Other Non-TIH/PIH Material‟ could potentially have, indicating that one or both of these factors should be weighted fairly high. While the general findings/results identified above were considered when estimating risk factor weights, no quantitative data from this report were identified as suitable for direct application to risk factor weights.

Base Case Risk Assessment: Data Analysis & Tests (Volpe Center) Overview: The Base Case Risk Assessment presentation [6] describes work performed by the Volpe Center in summarizing the “current level of risk” for a number of territories, as it summarizes results from various linear and anti-log multiple regression analyses that were performed. The study examines segments where PPAs have taken place, as well as other mainline railroad segments. The analyses described in this presentation are assumed to be a follow-on to the analyses documented in the RSAC report summarized in the previous subsection.

Risk Factors Addressed: The independent variables of the regression analyses are maximum allowable speed, average daily train count, and method of operation.

Key Findings/Results: The bulleted list below identifies findings of this study that appear to be relevant to the risk factors considered in the risk prioritization model.  Slide 7 indicates that annual PPA cost/train-mile was estimated to be 0.0544 for “Auto” territory, 0.0675 for TCS territory, 0.0590 for ABS territory, and 0.122 for dark territory. These values were calculated by dividing the annual PPA cost in territories of each method of operation by the annual train-miles for each method of operation.  Slide 46 concludes that “Both linear and anti-log multiple regression support the conjectures that the risk increases when the speed increases in all territories.”  Although train count is included as an independent variable in these analyses, the presentation slides contain no conclusions about this factor other than to say, on Slide 46, that risk is “associated with” train count.  Additional detailed quantitative findings are also included in this presentation, but it would be preferable to address a number of questions/concerns related to the analyses documented in this presentation before using these data. General Comments on How Results Were Used In Estimation of Risk Factor Weights: While an approach of the nature of that documented in this presentation would seem to be a reasonable approach for generating risk factor weights, additional information about the specific analyses documented in this presentation is needed before the findings of these analyses can be considered for use in estimating risk factor weights with a reasonable degree of confidence. While there were several questions that arose in the review of this report, only one of these questions is discussed below, as an example.

The coefficients of determination (R2), which are provided in the column of data under “RSQ” in the tables of results throughout this presentation, are extremely low for all of the regression models. The

177

PTCIP coefficient of determination is intended to provide an indication of the “goodness of fit”, as it measures the proportionate reduction of total variation in Y (which in these models was risk measured as PPA cost/train mile) associated with the use of the X variables in the model (method of operation, train count, speed). An R2 value, by definition, is constrained to be a value between 0 and 1, with a value equal to 1 indicating that the Y observations fall directly on the fitted regression surface (perfectly explaining the observed variation in Y) and a value equal to 0 indicating that the model does not at all explain the observed variation. The lowest R2 value of all of the regression models shown here is the 0.015% in the linear regression of risk for all territory (slides 25 and 35), while the highest R2 value of all of the models is the 1.3% in the anti-log regression of risk in “Auto” territory (slide 29). The presence of such low R2 values indicates that the model may not be well-suited to the data, and therefore, the results of this model were not taken into consideration when estimating risk factor weights. This lack of fit will impact both the significance and precision of the estimated parameters. Even if the parameters had suggested a slope that is consistent with theory, these slopes should not be considered to be validation of the theory as they are based upon a model that is not a strong fit to the data.

Risk and Train Control: A Framework for Analysis Overview: The report Risk and Train Control: A Framework for Analysis by Martland et al. [7] documents a PRA study which examined the effects of train control strategies on the risks of railroad operations. The research was initiated to address the question of how risk and the potential for risk reduction varies from one rail corridor to another, and it also attempts to assess how much reduction in risk can be expected from different train control strategies. The study considers a hypothetical corridor, which the report indicates was similar to the 183 actual corridors analyzed in the RSAC Report of the Railroad Safety Advisory Committee to the Federal Railroad Administrator: Implementation of Positive Train Control Systems [5], which was discussed above. The authors of this report described the study‟s level of analysis as “intermediate”, between highly detailed modeling of train behavior and crash analysis and the application of historical measures for accident rates and consequences. It should be noted that the cost of damage to equipment and track were not addressed in this study due to the authors‟ judgment that the costs from fatalities and injuries dominate these other costs.

Risk Factors Addressed: Factors addressed in this study include passenger traffic, traffic volume, number of tracks, train speed, grade, and curvature.

Key Findings/Results: The bulleted list below identifies findings/results of this study that appear to be relevant to the risk factors considered in the risk prioritization model. The report studies three different PTC technologies, identified as „PTC1‟, „PTC2‟, and „PTC3‟. When considering the results of this analysis, it was assumed that, of the three PTC technologies assessed, the PTC technology to be installed on CN‟s lines is closest to the „PTC2‟ technology; therefore, technology-specific results/findings presented here focus on analysis results pertaining to that technology.

The report states that “The PTC1 system is a bare-bones system that responds to signal overruns with full- service braking; it will also warn the crew and then stop the train if it exceeds speed limits… This unsophisticated system affords protection against signal overruns and overspeed accidents, but it responds slowly and clumsily to dangerous situations and does not use the digital communications link for further safety improvements.” It states that “The PTC2 system represents PTC systems that have been widely discussed and tested in recent years. The system is quicker in informing opposing trains that there is a problem, allowing them either to stop in time to prevent an accident or to slow down and reduce the severity of the accident… The on-board computer continually updates braking distances and controls the train‟s speed so that it does not exceed either its operating authorities or the speed limits… This system, like PTC1, still relies on the existing signal system to provide protection against broken rails.” In

178

PTCIP comparison, it notes that “The PTC3 system is more advanced, with features that can be imagined but that are not currently available (at the time of the study), including a new system for detection of broken rails.” Additional detail of these technologies are provided in that PRA report.

One thing to note when reviewing the summary of results below is the fact that, while dollar figures are shown, these dollar figures should not be interpreted as “real-world” PTC-related benefits. The dollar figures shown here are simply intended to provide a common metric for expressing relative risk and should not be taken out of this context.

 The first page of the report states that “Passenger traffic is the most important factor because the addition of passenger trains creates the possibility of catastrophic accidents with dozens of fatalities.”  The sensitivity study indicated that implementing „PTC2‟ technology on a corridor with twice as much passenger traffic as the base case corridor would lead to a reduction in risk greater than 2.8 times that achieved by implementing „PTC2‟ on the base case corridor (reductions of approximately $47.1M versus $16.9M, respectively).  The sensitivity study indicated that implementing „PTC2‟ technology on the base case corridor would lead to a reduction in risk greater than 4.2 times that achieved by implementing „PTC2‟ on a corridor with half as much total train volume as the base case corridor (reductions of approximately $16.9M versus $4M, respectively).  The report states that “As traffic volumes increase in terms of annual tonnage, railroads tend to operate heavier trains, and they tend to provide better track in order to be able to operate at higher speeds; railroads are very concerned about safety for the higher density lines. Statistical records of accidents per million train-miles may therefore show little or no relationship between traffic density and risk. However, when all other factors are held constant, risks are predicted to rise more than linearly with traffic volume.”  The sensitivity study indicated that implementing „PTC2‟ technology on a 1200-mile hypothetical corridor with all single track would lead to a reduction in risk greater than 4 times that achieved by implementing „PTC2‟ on a 1200-mile hypothetical corridor with double track (reductions of approximately $27.3M versus $6.8M, respectively).  The report states that “The risk of collision is much greater on single track than on multiple tracks.”  The report states that “…higher train speeds increase both the likelihood and the severity of the consequences of accidents if there is a signal overrun or a failure to obey a slow order.”  Although the report does not consider accidents involving hazardous materials, the authors do point out that “Accidents involving hazardous materials could conceivably be much more catastrophic even than a high-speed collision of two passenger trains.”  While the study did incorporate route characteristics such as grade and curvature in the modeling, the report does not address to what extent these characteristics had an effect on risk.

General Comments on How Results Were Used In Estimation of Risk Factor Weights: While this study seems to provide an indication of the extent to which different risk factors affect total risk under different methods of operation, it is acknowledged that various aspects of the CN subdivisions that are subject to the risk prioritization model vary from the segments that compose the hypothetical corridor assessed in this study. CN was mindful of differences such as these when considering these results during their estimation of risk factor weights. It was also acknowledged that the PRA and sensitivity analysis results presented in this study should only be considered directly applicable within the variable ranges

179

PTCIP assessed for the hypothetical corridor, such that extrapolation beyond these ranges is generally not recommended.

In many cases, the exact ranges of the variables considered in the model are unknown since they are not specified in the report. For example, the report states that “Various speeds for freight and passenger trains were established, with top speeds of 120 km/h (75 mph) for passenger trains and 60 km/h (37.5 mph) for freight trains,” and risk results are based on operations at these speeds. However, it is not known to what degree the speeds used in modeling the different segments of the hypothetical corridor varied from these “top speeds.” Also, while the sensitivity studies demonstrate how changes in speed of ± 6.2 mile/hour (mph) from the base case speeds affect risk, the base case speeds are not known and it is also not known how risk would be affected if speeds were varied beyond the 12.4 mph range studied here.

Keeping the various caveats above in mind, it is also true that the results of this PRA study are among the only available data points that are somewhat applicable to the risk factor weighting exercise, and while it may not be advisable to give too much weight to the results of this study, it does seem reasonable to consider how the risk factors rank with regard to the extent to which they affect risk, and it also seems reasonable to consider the results as “rough order of magnitude” estimates to be considered when assigning risk factor weights.

It should be noted that the risk factor weights may, to a certain degree, depend on the risk factor levels and vice versa. So, when assigning risk factor weights, the units that are used to quantify the risk factors, the range of values that compose the risk factor levels, and the magnitude of difference between risk factor levels are all things that can be taken into consideration. For example, the sensitivity study indicated that installing „PTC2‟ technology on a 1200-mile hypothetical corridor with all single track would lead to greater than $27M in risk reductions relative to the base case ABS corridor, while installing „PTC2‟ on a 1200-mile hypothetical corridor with double track would only achieve $6.8M in risk reductions relative to the base case ABS corridor. At first glance, this $20M difference in benefit might seem to indicate that the „Number of Tracks‟ risk factor then should be assigned a weight not far below that of „Presence and Volume of Passenger Traffic‟, but when taking into account the fact that the majority of CN subdivisions have zero miles of double track, with the average amount of double track for the CN subdivisions (of those that fit the Subpart I criteria for PTC installation) at approximately 6 miles, and the most mileage of double track, by far, being the approximately 55 miles on the C&M subdivision, it seems that the additional reduction in risk achieved by installing „PTC2‟ on single track versus double track may be relatively insignificant in CN‟s risk prioritization model.

In making a very rough approximation, it was assumed that the difference in „PTC2‟ benefit between single and double track increased linearly with train-miles, resulting in an approximately $1.39 difference in „PTC2‟ benefit per train-mile. Assuming an average train weight of 6000 tons, the MGT values and subdivision lengths were used to approximate average annual train-miles for CN subdivisions (considering only those to be equipped with PTC), which was found to be roughly 314,000 annual train- miles. Considering a subdivision of average train-miles, if 100% of the subdivision were single track, the average annual benefit could be approximated by multiplying 314,000 by $1.39, which results in a benefit differential of about $437K. Considering an average subdivision with 80% single track, the benefit differential would be approximately $350K, and taking the difference between the two levels of coverage results in a value of approximately $87K, which would provide the approximate difference in risk between average subdivisions in the factor levels defined in Section 3.4 below. This estimated difference in risk between single track and double track does not take into account the fact that the model in this PRA study did not seem to address other potential risks associated with multiple track, including the potential for secondary collisions on multiple track or increased risk of work zone-related accidents on multiple track, which might lessen the “importance” of this risk factor even further.

180

PTCIP

The study indicated that the „Presence and Volume of Passenger Traffic‟ is the most important risk factor. The weight assigned to this factor would, therefore, likely be higher than the weights assigned to the other risk factors addressed in this study. Limits of the risk factor levels for the „Presence and Volume of Passenger Traffic‟ risk factor were defined to be 2, 4, 8, and 16 passenger trains per day (see Section 3.2 below for discussion) in the risk prioritization model, such that the limit of each level is double that of the level below and half that of the level above. The sensitivity study indicated that implementing „PTC2‟ technology on a corridor with twice as much passenger traffic as the base case corridor (9.7 million train- miles vs. 4.9 million train-miles) would lead to a reduction in risk greater than 2.8 times that achieved by implementing „PTC2‟ on the base case corridor (reductions of approximately $47.1M versus $16.9M, respectively).

For the purposes of providing a very rough approximation, it was assumed that the differences in „PTC2‟ benefit varied linearly with train-miles of passenger traffic. The difference between $47.1M and $16.9M divided by the difference between 9.7 million train-miles and 4.9 million train-miles results in a value of $6.29 difference in benefit per passenger train-mile annually. Taking the average annual passenger train- miles for all CN subdivisions that have passenger traffic (1,012,948 passenger train miles/7 subdivisions =144,706 passenger train miles) and multiplying this value by $6.29 results in a value of $910K, which represents „PTC2‟ benefit for the average subdivision with passenger traffic. This value of $910K was used to provide a very rough approximation of the average annual difference in risk benefit achieved between the various „Presence and Volume of Passenger Traffic‟ risk factor levels, and it was compared to values calculated for other risk factor results from this study. For example, using this very coarse approximation, the $910K was compared to CN‟s $87K average annual additional benefit of installing PTC on single track versus double track, indicating that „Presence and Volume of Passenger Traffic‟ should receive a weight roughly an order of magnitude higher than „Number of Tracks‟ risk factor.

The sensitivity study indicated that implementing „PTC2‟ technology on the base case corridor would lead to a reduction in risk greater than 4.2 times that achieved by implementing „PTC2‟ on a corridor with half as much total train volume as the base case corridor. This would result in risk reductions of approximately $16.9M versus $3.96M, for corridors with 14.5 million train-miles and 7.25 million train- miles, respectively. While CN‟s risk prioritization model assesses the „Annual MGT level‟ risk factor separately from „Presence and volume of passenger traffic‟, such that the „Annual MGT level‟ risk factor does not include volume contributed by passenger traffic, the total train volume assessed in the PRA study reflects a combination of freight and passenger traffic at a ratio of 2-to-1. The fact that passenger traffic is included in this total volume may have somewhat skewed the calculations performed here, but the traffic mix ratio appears to be constant in the study‟s sensitivity analysis, thereby reducing additional variability. Again assuming the difference in „PTC2‟ benefit is linear, a very rough approximation of $1.78 in additional „PTC2‟ benefit per train-mile is achieved when doubling the volume of traffic.

On average, the difference between „Annual MGT Level‟ risk factor levels is a factor of approximately 1.7, with the upper limit of factor levels equal to 5, 10, 20, 30, and 40 MGTs (see Section 3.1 below). Again, extrapolating using the $1.78 average additional benefit per train-mile based on doubling MGT levels approximates a value of $1.51, assuming an average factor of 1.7 between risk factor levels. Multiplying 314,000 train-miles (calculated above) by the $1.51 average additional benefit per train-mile resulted in an estimate of roughly $474K additional difference between risk factor levels. This is just over half of the $910K value approximated for „Presence and Volume of Passenger Traffic‟. Based on the very rough calculations documented here, the risk factor weight assigned to „Annual MGT Level‟ was found to be just over half of that assigned to „Presence and Volume of Passenger Traffic‟ and over five times that assigned to „Number of Tracks‟.

Similar approximations were made for the „Speed of Operations‟ risk factor. According to the results of the sensitivity study, implementing „PTC2‟ on corridors with an approximately 12.4 mph difference in operating speed (one corridor 6.2 mph above the base case operating speed, and the other corridor

181

PTCIP

6.2 mph below), a $27.5M reduction in risk per 14.5 million train-miles is achieved at the higher speed while a reduction of $10.2M per 14.5 million train-miles is achieved at the lower speed. Making a rough approximation that reduction in risk varies linearly with increases in speed, a value of $1.38M benefit per mph is found over 14.15 million train-miles. An average difference of 10 mph exists between the risk factor levels defined for the „Speed of Operations‟ risk factor in Section 3.6 below. Multiplying this average difference of 0 mph by the $1.38M benefit per mph results in a value of $13.8M difference between risk factor levels, assuming 14.5 million train-miles. Again, for the purposes of rough approximation, it was assumed that benefit per mph varies linearly with train-miles, such that the average 314,000 annual train-mile subdivision would experience a benefit of $300,000 between risk factor levels for „Speed of Operations‟. This means that „Speed of Operations‟ was roughly one-third the weight of „Presence and Volume of Passenger Traffic‟, over three times the weight of „Number of Tracks‟, and just over 50% less than „Annual MGT Level‟. Figure A-1 below shows the relative weights of the four risk factors approximated here. These approximations are also briefly discussed in Section 3.0 under the various risk factors to which these estimates are applicable. The approximations were just one set of inputs to be considered when assigning risk factor weights and were interpreted simply as rough approximations that were made using available information.

Figure A-1. Approximations for Consideration in Estimation of Risk Factor Weights

182

PTCIP

Appendix D: Interoperability Letters of Understanding

183

PTCIP

184

PTCIP

185

PTCIP

186

PTCIP

187

PTCIP

188

PTCIP

189

PTCIP

190

PTCIP

191

PTCIP

192

PTCIP

193

PTCIP

END OF DOCUMENT

194