Stirling City Centre Light Rail Feasibility Study

- Phase 2

November 2010

Public Transport Authority of WA

Parsons Brinckerhoff Australia Pty Limited ABN 80 078 004 798 Level 5 503 Murray Street PERTH WA 6000 PO Box 7181 CLOISTERS SQUARE WA 6850 Australia Telephone +61 8 9489 9700 Facsimile +61 8 9489 9777 Email [email protected]

Certified to ISO 9001, ISO 14001, AS/NZS 4801 10-0477-02-2106689A A+ GRI Rating: Sustainability Report 2009

Revision Details Date Amended By 00 Original 01 Revision A 02 Revision B 24 Nov. 10 B. McMahon

©Parsons Brinckerhoff Australia Pty Limited (PB) [2010]. Copyright in the drawings, information and data recorded in this document (the information) is the property of PB. This document and the information are solely for the use of the authorised recipient and this document may not be used, copied or reproduced in whole or part for any purpose other than that for which it was supplied by PB. PB makes no representation, undertakes no duty and accepts no responsibility to any third party who may use or rely upon this document or the information. Author: Brian McMahon ......

Signed: ......

Reviewer: Dick Fleming ......

Signed: ......

Approved by: Dick Fleming ......

Signed: ......

Date: 24 November 2010 ......

Distribution: ......

Please note that when viewed electronically this document may contain pages that have been intentionally left blank. These blank pages may occur because in consideration of the environment and for your convenience, this document has been set up so that it can be printed correctly in double-sided format.

10-0477-02-2106689A Stirling City Centre Light Rail Feasibility Study - Phase 2

Contents

Page number

Executive summary v

1. Introduction 1

1.1 Background 1 1.2 Modal comparison matrix 3 1.3 Phase 1 results 4 1.4 System requirements 6 1.4.1 Operational 6 1.4.2 Environmental 7 1.4.3 Economic 8

2. Modal characteristics 9

2.1 Bus on street 9 2.1.1 Patronage capacity 10 2.1.2 Capital costs 10 2.1.3 Operational costs 11 2.1.4 Value uplift 11 2.1.5 Running ways 11 2.1.6 Corridor reservations 11 2.1.7 Stations 11 2.1.8 Vehicles 12 2.1.9 Other elements 12 2.2 Tram on street 13 2.2.1 Patronage capacity 14 2.2.2 Capital costs 14 2.2.3 Operation costs 15 2.2.4 Value uplift 15 2.2.5 Running ways 15 2.2.6 Corridor reservations 16 2.2.7 Stations 16 2.2.8 Vehicles 17 2.2.9 Other elements 17 2.3 Bus Rapid Transit (BRT) 18 2.3.1 Patronage capacity 19 2.3.2 Capital costs 19 2.3.3 Operational costs 20 2.3.4 Value uplift 20 2.3.5 Running ways 20 2.3.6 Corridor reservations 21 2.3.7 Stations 22 2.3.8 Vehicles 22 2.3.9 Other elements 23

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page i

Stirling City Centre Light Rail Feasibility Study - Phase 2

Contents (Continued) Page number 2.4 Light Rail Transit (LRT) 24 2.4.1 Patronage capacity 25 2.4.2 Construction costs 25 2.4.3 Operation costs 26 2.4.4 Value uplift 26 2.4.5 Running ways 26 2.4.6 Corridor reservation 26 2.4.7 Stations 26 2.4.8 Vehicles 27 2.4.9 2.4.9 Other elements 27 2.5 Emerging technology – trams on tyres 28 2.5.1 Patronage capacity 28 2.5.2 Capital costs 28 2.5.3 Operational costs 29 2.5.4 Value uplift 29 2.5.5 Running way 29 2.5.6 Corridor requirements 30 2.5.7 Stations 30 2.5.8 Vehicles 30 2.5.9 Other elements 30

3. Value capture 33

3.1 Joint development 34 3.2 Benefitted areas charges 34 3.3 Tax Increment Financing 35 3.4 Revenue sharing 36 3.5 User fees 36 3.6 Other Innovations 36

4. Refined patronage forecasts – TODTrips model 39

4.1 Background 39 4.2 TODTrips model 39 4.3 Assessment result of five transit modal scenarios 41 4.3.1 Scenario 1 – Base case (Bus) 42 4.3.2 Scenario 2 – Street Car/Tram 43 4.3.3 Scenario 3A – LRT 44 4.3.4 Scenario 3B – BRT 45 4.3.5 Scenario 4 – LRT (Single Sided) 45 4.4 Internal trips – mode share and ridership estimates 46 4.5 External trips – mode share and ridership estimates 48 4.6 Combined internal and external trips - mode share and ridership estimates 49

5. Summary 51

Page ii 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

Contents (Continued)

Page number

List of tables Page number

Table 1.1 Summary of findings by scenario 5 Table 2.1 Typical Conventional Bus system characteristics 10 Table 2.2 Typical Street Car/Tram characteristics 13 Table 2.3 Adelaide Street Car/Tram characteristics 14 Table 2.4 Range of BRT characteristics 18 Table 2.5 Typical BRT characteristics 19 Table 2.6 Existing BRT characteristics 20 Table 2.7 Typical LRT characteristics 24 Table 2.8 Existing LRT system characteristics 25 Table 2.9 Existing TransLohr systems - Costs 29 Table 4.1 Existing public transports operating environment for Stirling 41 Table 4.2 Scenario base (Bus’s) operating environment 42 Table 4.3 Scenario S2 (Street Car/Trams) operating environment 43 Table 4.4 Scenario 3A (LRT’s) operating environment 44 Table 4.5 Scenario 3B (BRT’s) operating environment 45 Table 4.6 Scenario 4 (LRT to one side) operating environment 46 Table 4.7 Low car use – Mode share for internal trips 47 Table 4.8 High car use scenario – Mode share of internal trips 47 Table 4.9 Low car use scenario – ridership share among different transport modes for internal trips 47 Table 4.10 High car use scenario – ridership share among different transport modes for internal trips 48 Table 4.11 Distribution of regional access and egress by external trips into and out of study area (I-E and E-I movements) 48 Table 4.12 Modal split of external trips using local transit services 49 Table 4.13 Ridership estimates of external trips (I-E and E-I movements) using transit services 49 Table 4.14 Low car use – mode share for combined internal and external trips 50 Table 4.15 High car use – mode share for combined internal and external trips 50 Table 4.16 Low car use – ridership estimates for combined internal and external trips 50 Table 4.17 High car use – ridership estimates for combined internal and external trips 50

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page iii

Stirling City Centre Light Rail Feasibility Study - Phase 2

Contents (Continued)

Page number

List of figures Page number

Figure 1.1 Stirling study area 2 Figure 1.2 Essential traits of a successful short haul transit service 6 Figure 2.1 Centre median running LRT with outside platforms in Portland, Oregon 24 Figure 4.1 Stop locations of existing public transport service 41 Figure 4.2 Stop locations of new local bus service in Scenario 1 42 Figure 4.3 Stop locations of Street Car service in Scenario 2 43 Figure 4.4 Stop locations of LRT service in Scenario 3A 44 Figure 4.5 Stop locations of BRT service in Scenario 3B 45 Figure 4.6 Stop locations of LRT (to one side) service in Scenario 4 46

List of photographs Page number

Photo 2.1 Conventional bus 9 Photo 2.2 Streetcar (tram) in San Francisco with parallel bike lane and on-street parking 13 Photo 2.3 Centre island platform along Eugene Oregon EmX BRT line 18 Photo 2.4 Fully dedicated BRT corridor 21

Appendices

Appendix A Mode comparison summary Appendix B Operating scenario and costs Appendix C Comparative operating characteristics Appendix D Stirling City Centre- Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Page iv 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

Executive summary

The Stirling City Centre Alliance (SCCA) is seeking to create a high quality transit corridor, in particular light rail, as a significant feature in the city centre’s revitalization. The SCCA has initiated an investigation into the feasibility of light rail serving the Stirling City Centre (SCC) and Scarborough Beach Road linked to the Glendalough rail station in Osborne Park. In this phase (Phase 2) of the feasibility assessment, the following tasks were conducted, the results of which are presented in this report.

 compare the operating characteristics of the ‘high quality transit’ modal options and running way environments in general based on a desk-top assessment

 refine route and running way options for the study area

 refine the patronage forecasts for the study area by mode using the PB TODTrips model

 assess the high level costs and potential value-uplift of the modal options for the study area

 identify a preferred alignment, mode and running way environment for the study area.

In addition, the report summarizes potential value capture benefits and techniques that could be used to help finance the proposed infrastructure.

The findings of the analysis supports the further consideration of quality transit in the study area. As part of the desk top assessment of modal options, five different forms of ‘high quality’ mass transit systems have been examined. The different modes include standard buses on street, trams on street, Bus Rapid Transit (BRT) with buses in exclusive lanes with high priority and Light Rail Transit (LRT) in exclusive lanes with priority. An additional emerging technology, tram on tyres has been assessed (TransLohr).

Five scenarios were setup in TODTrips to represent alternative modal options and operating environments that could be considered to serve the SCC in 2031. The five alternative modal scenarios analysed included:

 Base case with 2031 bus option

 Streetcar/Tram on kerbside

 LRT in dedicated centre median

 BRT in dedicated centre median

 LRT in dedicated single sided running way.

Due to the proprietary nature of the TransLohr technology, which would limit the project sponsors to a single manufacturer, this mode was not considered in patronage forecasts. In general, the broad operating environment for each scenario was set up to maintain appropriate existing public transport services with the addition of a new mode with a specified level of service.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page v

Stirling City Centre Light Rail Feasibility Study - Phase 2

The potential high quality transit system is part of an integrated land use and transport strategy for Stirling City Centre and the Glendalough/Osborne Park redevelopment areas. The potential transit upgrade is inextricably linked to a robust land use vision, a parking strategy to reduce private motor vehicle usage, a parking levy to fund catalytic transport infrastructure, control of roadway capacity to avoid recongestion of Scarborough Beach Road and a comprehensive bicycling and pedestrian strategy. Each of the components is mutually supportive and, as a whole, creates a unique and supportive environment for altering travel behaviour.

In Phase 1, the base case analysis showed an estimated light rail patronage of approximately 27,000 trips on an average weekday. To place these findings in context, comparison was made with light rail systems introduced in recent years in the United States. Comparison with these figures indicates that the Stirling light rail system is definitely ‘in the ballpark’.

Increasing development to the higher ‘aspirational’ levels would increase this somewhat to approximately 31,000 trips, while if the transit mode increased from 5.5% to 15% as many as 41,000 trips per day might be expected. (Note: The Phase 2 assessment assumed slightly higher mode shares by mode as result of the parking and cycling strategies that have been advanced by the City of Stirling subsequent to the Phase 1 study to support the integrated land use and transport strategy).

The analysis considered the potential corridor between Stirling Station and Glendalough Station as two stages. Stage One comprised a north south corridor along a realigned Ellen Stirling Boulevard or Stephenson Avenue. This stage would almost certainly not be justified on patronage grounds alone over the short term. However as a development catalyst it displays some merit. Stage Two included an east west corridor along Scarborough Beach Road between Ellen Stirling Boulevard and Glendalough Station.

The Phase 1 study suggested that Stages One and Two together would probably generate significant levels of associated development and patronage provided that there is ‘buy in’ from land holders and developers in the corridor.

Alternatively, Stage One of the line should be used as a catalyst for development within the Stirling Central area to help encourage the preferred patterns of development. In this role, Stage One must be tied to commitments to develop transit supportive land uses within an acceptable timeframe and firm agreements should be in place to adequately cover operating costs. In addition, under this scenario, Stage One should only proceed if there is certainty that the full system will be built to ensure a more financially sustainable outcome.

The Phase 2 study reinforces the findings that a high quality transit system, such as a LRT or street running tram is viable if supporting land use and transport policies are in place.

Page vi 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

The following conclusions and observations are made in relation to the results of the modelling:

 There is a potentially strong market for a high quality transit system to provide for travel within the study area and to facilitate the use of public transport for access to the area from other parts of the metropolitan area. In particular, there is strong potential for the operation of an effective internal transit system in the Stirling Centre. In other words, the transit system will function well as a ‘pedestrian accelerator’ for relatively short trips amongst origins and destinations within the study area.

 The modelling results show that demand could be in the range of around 40,000 to 55,000 passengers per day. This result is considered to be relatively high and has been driven by the land use assumptions and the overall high level of development included in the model. These figures should be reviewed as part of a practical assessment of the development potential in the study area.

 Further refinement of the patronage estimate should be completed to inform the system design. This should include an assessment of an initial operating segment on the north-south alignment only (Phase 1), a base scenario using current development conditions, higher transit mode shares in Phase 1 alone (e.g., 15%), and with alternative local bus operating conditions (e.g., the 400 series bus operating more frequently)

 The transit system has a strong role to play in minimising the use of private motor vehicles for movement within the centre and minimising the demand for parking.

 With regards the modes tested, the street running transit (tram) shows that it has the potential to attract marginally more passengers than the other options. This mode is the most accessible, with the highest number of stops which underlines the importance of selecting a mode which can be closely integrated with development along the corridor.

 The land use development forecasts herein for the Bus and BRT scenarios are equivalent to those for the LRT and tram scenarios. However, in reality, these thresholds of development would likely not be attained since developers would not be attracted to invest in the corridor under the Bus and BRT scenarios. Development tends to ‘follow the rails’ with a notable preference for investment along rail corridors due to the increased property values and returns. As a result, forecasting should be refined to reflect a lower level of development for the bus-based scenarios.

 The design of the transit system, the final decision regarding the streets in which it will operate and the delivery of developments which support active street frontages will have a strong bearing on the ultimate success of the transit system.

 It is essential that supportive land use and transportation policy framework be in place prior to implementation to allow the catalytic effects and economic uplift to be produced effectively and to allow for value capture.

 Close integration is required at Stirling and Glendalough Railway Stations to ensure barrier free seamless interchange conditions for passengers to maximise the attractiveness of the transit system for people travelling from outside the study area. In addition, the Stirling transit system should be included as part of the Perth PTA fare system to ensure that riders get a cost penalty free transfer from rail and bus to the new Transit system. Lastly, the Stirling system will require integration and standardization with the overall Perth LRT system as set forth in the 20 Year Transit Plan (pending).

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page vii

Stirling City Centre Light Rail Feasibility Study - Phase 2

Based on the modelling findings and the potential land use integration and transport characteristics, a hybrid tram/LRT system is recommended for further consideration in the next phase (i.e. Concept Design and Final Feasibility). The hybrid would include a centre median dedicated LRT (and potential BRT) along Scarborough Beach Road. This running environment would maintain operational reliability by avoiding congested travel lanes. It is recommended that mid-block traffic signals be introduced along Scarborough Beach Road to allow for two or more additional stations and safe pedestrian and cyclist access to be included along the corridor. As shown in the modelling, the additional stations allowed by the streetcar/tram served to increase patronage.

The hybrid would include either a streetcar along a realigned Ellen Stirling Boulevard or a single side running LRT along the west side of Stephenson Avenue. The benefit of the former is the inclusion of additional stations and better integration with supportive, surrounding land uses. The benefit of the latter is the placement of the stations in closer and more direct walking access to the land uses due to separation created by the day-lighted stream and parklands on the east side of the street.

In terms of value capture, it is clear that significant economic and property value uplift occurs with the introduction of light rail and street cars if supportive conditions are in place. These include a general growth in the real estate market and the presence of congestion so that value is attached to the presence of the high quality transit as a more reliable and convenient alternative to other modes. There is also growing evidence that BRT systems can provide similar benefits however the level of documentation is not as extensive.

A modal comparison summary matrix is provided in Appendix A to illustrate characteristics of existing high quality transit systems globally. In addition, a range of hypothetical operating characteristics, scenarios and costs are presented in Appendix B for each mode. The matrices have been developed to allow for the assessment of optimal service plans based on the incremental growth of patronage over time, required equipment and optimal frequency of service. The matrices should be used to further refine the service plan as part of the next steps.

The hypothetical costs included in Appendix B for each mode include capital and operating costs as well as life cycle costs for various operating scenarios based on potential patronage and service patterns. One noteworthy observation based on the life cycle cost analysis is that trams and light rail appear to have lower long term costs than bus based systems. However, verification of this observation will require additional refinement based on the actual proposed operating plan and concept design in Phase 3.

The reader is forewarned not to make generalizations about the performance of the Stirling system based on international and national averages as presented in the report and in Appendices A and B. Costs and performance measures are based on averages from urban or suburban settings that may not be relevant to the Stirling corridor. The information simply provides some basic parameters for illustrative purposes. A more detailed concept design, service plan and final patronage forecast is recommended as a next step (Phase 3) to more precisely determine costs for the Stirling LRT.

Page viii 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

1. Introduction

1.1 Background

The Stirling City Centre Alliance (SCCA) is seeking to create a high quality transit corridor, in particular light rail, as a significant feature in the city centre’s revitalization. The SCCA has initiated an investigation into the feasibility of light rail serving the Stirling City Centre and Scarborough Beach Road. In this phase (Phase 2) of the feasibility assessment, the following tasks were conducted, the results of which are presented in this report.

 compare the operating characteristics of the ‘high quality transit’ modal options and running way environments in general based on a desk-top assessment

 refine route and running way options for the study area

 refine the patronage forecasts for the study area by mode using the TODTrips model

 assess the high level costs and potential value-uplift of the modal options for the study area

 identify a preferred alignment, mode and running way environment for the study area.

In addition, the report summarizes potential value capture benefits and techniques that could be used to help finance the proposed infrastructure.

Phase 1 elements are summarised in Section 1.3 below.

As shown in Figure 1.1, the study corridor is the approximately 3.4 kilometre corridor extending in approximately an L-shape between the Stirling and Glendalough Stations on the Perth Northern Rail Line in the City of Stirling. The proposed light rail corridor would extend in a north south direction between the Stirling Station and Scarborough Beach Road, through the Stirling City Centre, along either existing realigned Ellen Stirling Boulevard or the proposed Stephenson Boulevard. From this location, the east west portion of the line would extend to Glendalough Station along Scarborough Beach Road.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 1

Stirling City Centre Light Rail Feasibility Study - Phase 2

Figure 1.1 Stirling study area

This study has been triggered by the desire of the SCCA to redevelop the Stirling City Centre into a major urban centre incorporating significant new commercial, retail, community and residential opportunities. Studies are also underway to revitalize and intensify the Scarborough Beach Road corridor through the Herdsman Lake Business Park and Glendalough. The redevelopment would represent a fundamental change in the nature of the area from one dominated by a major shopping mall, warehouses and bulky goods retail to a city centre featuring 24/7 activity.

Previous work undertaken by the SCCA indicated that the road network would not be capable of handling the proposed increase in activity without a major shift to other transport modes, including public transport. An LRT has been proposed as one method of accommodating the required transport demand while at the same time providing an ‘anchor’ to encourage the type and level of development intended.

Page 2 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

It is expected by the project sponsors that a high quality transit system could catalyse the desired redevelopment and infill in the study area. It is also anticipated that the quality transit could generate higher real estate market values that could be ‘captured’ by local and state government to finance a portion of the infrastructure to support this redevelopment. An additional key role of the transit corridor will be to link people to the rail system at Stirling and Glendalough Stations. It will be critical to the success of the transport system in meeting the mode share targets for the Regional Centre that this ‘short haul’ distributor function work effectively by providing a seamless connection to the rail and bus networks. In turn the rail and bus networks will provide the links to the wider metropolitan area. Given the fact that the majority of people will be accessing destinations which are outside reasonable walking distance of the railway stations quality transit is well suited to serving this distributor role.

As part of the desk top assessment of modal options, five different forms of ‘high quality’ mass transit systems have been examined. The different modes include standard buses on street, trams on street, buses in exclusive lanes with high priority and trams/LRT in exclusive lanes with priority. An additional emerging technology, tram on tyres has been assessed. The five modes have been categorised as the following:

 Bus Bus on street without priority

 Bus Rapid Transit (BRT) Bus in exclusive lanes with priority

 Streetcar Rail base tram on street without priority

 Light Rail Transit (LRT) Light rail in exclusive lanes with priority

 Emerging Technology Tyre base tram without priority.

Chapter 2 describes the operating characteristics of each mode. Section 1.2 below summarizes the elements addressed in the modal comparison.

Chapter 3 sets forth a summary the potential tools to capture the economic benefits associated with fixed transit infrastructure.

The hypothetical system operating characteristics and refined patronage forecasts for each mode (excepting TransLohr) based on the TODTrips analysis are presented in Chapter 4.

Chapter 5 summarizes the key findings and next steps for the introduction of LRT to Stirling.

1.2 Modal comparison matrix

Based on the results of the desk top assessment, Appendix A presents a side by side comparison of the five modes by each of the following characteristics:

 transit mode use in other cities

 capacity by vehicle type – different vehicle type by seated, standees and total passengers numbers

 peak hour capacity using international examples and an estimate of the theoretical capacity by using set frequencies and the research vehicle type

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 3

Stirling City Centre Light Rail Feasibility Study - Phase 2

 capital expenditure – cost per km for infrastructure using international examples of infrastructure and construction costs per kilometre in 2009/2010 Australian dollars

 cost per vehicle – international examples of average cost for vehicles in 2009/2010 Australian dollars

 operating costs – estimated by kilometre, revenue passenger kilometre or vehicle hour

 operating speeds (Average and Maximums) – approximate timetabled running speed and the maximum achievable speed by type of vehicle

 turning radii

 power source

 timetable and technology reliability – ability for the technology to maintain timetabled running times

 technology maturity – level of maturity that the technology is at in terms of reliability and use in other cities

 integration with pedestrian realm and land uses – how well the technology integrates with the pedestrian environment and surrounding land uses

 visual amenity – Issues that the technology has on the surrounding area

 value uplift and redevelopment catalyst – influence the technology has on property values and an effect it has on increasing or encouraging development around the corridor and stations.

1.3 Phase 1 results

Phase 1 of the Stirling Centre Light Rail Feasibility Study undertook a ‘high level’ examination of potential patronage of a light rail system to support the Stirling – Osborne Park corridor. This was not a detailed modelling exercise but rather a broad ‘spreadsheet’ modelling approach with the prime objective of establishing if the proposed level of land use intensity could generate sufficient demand to support a light rail system to warrant moving to a more detailed study of potions. In Phase 1, the base case analysis showed an estimated light rail patronage of approximately 27,000 trips on an average weekday. Increasing development to the higher ‘aspirational’ levels would increase this somewhat to approximately 31,000 trips, while if the transit mode increased from 5% to 15% as many as 41,000 trips per day might be expected.

The high level analysis in this phase was based on outputs from the Department of Planning’s STEM model for 2031. It also considered Scarborough Beach Road Population and Land Use Study (Syme Marmion 2009). The base case relied on mid-range development projections from this study, resulting in a combined forecast a resident population of 23,500 and an employee population of 36,650. The base case also assumed a public transport mode share of 5.5% that is current for Greater Perth.

Page 4 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

These figures were found to be very sensitive to changes in shopping related trips and trip behaviour, as was shown by the test results of reducing the retail trip generation rate for the Osborne Park area. It is worth noting that the patronage estimation included in the Syme Marmion report likely underestimates retail trips significantly, as their estimate considers only trips by householders and inbound work trips.

The analysis also considered the impacts on ridership of phasing the construction into two phases. In Phase 1, the line would extend north south for approximately 1 kilometre between Stirling Station and south of Scarborough Beach Road. The importance of future development along Scarborough Beach Road in Osborne Park and Glendalough areas was highlighted by the testing of a Phase 1 tram route only. This returned a much lower estimate of patronage (around 6,000) trips compared to the Base Case.

The findings of each scenario are summarised in Table 1.1 below.

Table 1.1 Summary of findings by scenario

Reduced Phase 1 Improved PT Increased retail Base case tram/LRT mode share development intensity only (15%) (O.P.) Total trips 479,000 479,000 598,000 479,000 396,000 Tram/LRT trips 27,000 6,000 31,000 41,000 21,000

To place these findings in context, comparison was made with light rail systems introduced in recent years in the United States). Comparison with these figures indicates that the Stirling light rail system is definitely ‘in the ballpark’.

It was further concluded that an investigation should be conducted to refine the initial operating segment or extent of Stage One. Minor changes in the segment may benefit the ridership by linking existing or near term destinations, a critical characteristic of successful LRTs. In other words, the initial operating segment of the LRT should connect ‘somewhere’ to ‘somewhere’ to foster ridership and to catalyse development. The segment should not be based on right of way availability as the primary factor, but on the connection of destinations and the frontage on development sites.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 5

Stirling City Centre Light Rail Feasibility Study - Phase 2

On these results, further detailed study of the proposed LRT appeared warranted. Thus Phase 2 of the study was undertaken. It was recommended in Phase 1 that the ability of an LRT to shape and support the type, level and location of urban development desired be considered in Phase 2. The requirements for effective integration with the rail and bus networks (both operationally and physically) and potential issues relating spatial needs and co-location of with other modes were also recommended for consideration in Phase 2.

1.4 System requirements

In order to assess the general characteristics of each mode, the following system operational, environmental and economic requirements have been identified. Many of the factors cannot be quantified or measured in this analysis due to the study scope. However, the factors were taken into consideration in developing the alternative modal and running way concepts.

1.4.1 Operational

 Generate and Attract Local Trips – the proposed transit system, including both the mode and service plan attracts riders with trips within the study area or to and from the study area. Less emphasis is given to trips travelling through the study area.

 Supports Desired Activities – the aim of the system is encourage an urban vibrancy along its streets. Desired activities include sitting, walking, biking and other daily activities.

 Short Haul Service – the primary purpose of the 3.4-kilometre line examined herein is to serve local as opposed to regional trips. Regional access to other centres is envisioned to be provided by the Northern Rail Line and the Circle Line route in the foreseeable future. Thus, a high frequency service with low waiting times is more important than speed of operation.

Figure 1.2 Essential traits of a successful short haul transit service

Page 6 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

 Priority Where Needed – In its role as a short haul service, priority in a dedicated right of way is only required where substantial congestion will occur throughout the day.

 Integration with PTA Long Haul Service – There are two other potential long term transit improvements that may require a dedicated right of way along Scarborough Beach Road. A dedicated right of way is sought by the PTA for a potential BRT line extending from Scarborough Beach through the corridor to the CBD. PTA has indicated its intent to operate a service with 3 to 5 minute headways along the roadway and is interested in co-locating the service with the short-haul service in a dedicated right of way. The PTA is also interested in a dedicated right of way for operation of a regional LRT connection from the study area south through Subiaco to the University of Western Australia. This concept is set forth as a year 11–20 horizon project in the 20 Year Transit Plan. In addition to integration with the proposed long haul BRT and LRT, the short haul service will need to integrates with the PTA’s conventional bus service.

 Meets Ultimate Passenger Demand – The Phase I forecast broadly estimates between 26,000–40,000 passengers per day. The Phase II forecasts resulting from the TODTrips modelling show the following range:

 Bus in Street

 Tram in Street

 BRT in Dedicated Right of Way

 LRT in Dedicated Right of Way.

 Due to private property ownership and the lack of access to parallel east west alignments, it is assumed that the east west segment will run along Scarborough Beach Road (SBR). It is further assumed that the proposed running way for the transit service should support the ultimate expansion of the right of way along Scarborough Beach Road from 30-metres to 42-metres. Two options, Ellen Stirling Boulevard and the proposed Stephenson Boulevard are considered for the north-south segment.

1.4.2 Environmental

 Carbon Reduction – ability to attract infill development will support savings in carbon dioxide.

 Low Traffic Impact – at a high level, the concept should support travel route choice for all modes including vehicular traffic. For example, the dedicated right of way should allow cross traffic at all intersections.

 Supports Constrained Parking Supply – since the study area cannot support additional car-based trips, parking will be constrained. The concept (including service plan) should attract riders to alleviate demand for parking and car-based trips.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 7

Stirling City Centre Light Rail Feasibility Study - Phase 2

1.4.3 Economic

 While more detailed economic cost and benefit analyses would be developed as part of a business case, the following economic factors were taken into consideration at a high level.

 appropriateness of Capital Cost for Benefits Delivered

 reasonableness of Operating Costs

 sensibility of Life Cycle Costs

 potential for Market Value Uplift and Land Use Activation.

Page 8 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

2. Modal characteristics

The following chapter presents the performance characteristics of the four modal alternatives. The intent is to introduce the modes in the context of the required characteristics for the short-haul Stirling Glendalough line. No single type of quality transit system is appropriate for all applications. The potential solution for Stirling should be determined based on an objective and comprehensive analysis of the alternatives. The criteria used in this analysis reflect the stated goals of the SCCA and these criteria may differ from those in other settings, even within Greater Perth.

A mode comparison summary matrix is provided in Appendix A. In addition, a range of hypothetical operating characteristics, scenarios and costs are presented in Appendix B for each mode. The costs include capital and operating costs as well as life cycle costs for various operating scenarios based on potential patronage and service patterns. The latter determines ultimate equipment needs.

The reader is forewarned not to make generalizations about the performance of the Stirling system based on international and national averages as presented below and in Appendices A and B. Costs and performance measures are based on averages from urban or suburban settings that may not be relevant to the Stirling corridor. The information simply provides some basic parameters for illustrative purposes. A more detailed concept design and service plan will need to be developed to more precisely determine costs for the Stirling LRT.

2.1 Bus on street

Buses are flexible, comparatively cheap to operate and relatively easy to implement on city streets. They can provide a high level of service and if marketed correctly can help alleviate congestion and increase the mode split between private vehicles and public transport. Buses can be implemented on a corridor within a short period and can are flexible enough that changes to running times, frequency of service and capacity can be implemented at short notice (provided that the vehicles are

available). Buses are able to provide the basic public transport function in almost Photo 2.1 Conventional bus all applications and are the most common form of transit mode across Australia and the world. However, in terms of attracting significant percentages of people away from private vehicles, standard buses in regular streets do not compare to other systems with higher priority and/or perceived public image. Without bus priority measures travelling times can be slow and journey times long. Close spacing between stops and unpredictable traffic conditions can reduce the reliability of the service and significantly slow down journey times. Circuitous routes are generally provided to users with a community type service rather than a dedicated mass transit service. Therefore, the appeal of buses over other transit modes, particularly for

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 9

Stirling City Centre Light Rail Feasibility Study - Phase 2

attracting ‘choice riders’ is considered to be less. Choice riders are those passengers who can afford to use public transport as opposed to those who need to for physical or financial reasons.

The average passenger trip length in kilometres for buses in the United States is (American Public Transportation Fact Book, 2010) is 6.25-kilometres. This reflects that buses operate primarily in central areas where origins and destinations are closer together.

Table 2.1 Typical Conventional Bus system characteristics

Transit type Street transit Typical Maximum Passengers per Hour 4,000 Max. Frequency (vehicle /hr) 60–80 Avg. Passenger Trip Length (km) 6.25 Typical Station Spacing (m) 250–400 Propulsion Energy Diesel, Euro-Diesel, Natural Gas, Biodiesel, Hydrogen, Hybrid Environmental Considerations Noise and street congestion Technological Maturity High

2.1.1 Patronage capacity

Standard buses are able to accommodate anywhere from 20 to 110 passenger per vehicle depending on size and configuration. From a basic level of service, standard buses are able to provide peak hour capacity of up to 6,600 passengers per direction, however, this is assuming that services are conducted using articulated buses and operate every minute. Although this may be possible theoretically, in a realistic situation, the actual capacity would be slightly lower.

For standard operations in peak hour conditions, buses without significant priority generally cannot keep to scheduled timetables. Thus when service levels fall below 5 minutes headways, bunching of services often occurs which results in one bus being overloaded and running late, while the second bus trailing behind the first with minimal numbers of passengers. Operating standard buses at one or two minute frequencies using standard infrastructure also causes issues with passenger loading, bus queuing, road congestion as well as stop lengths and infrastructure requirements. When the level of service reaches this degree, the optimum function of the service degrades.

2.1.2 Capital costs

Since standard buses operate on regular roadways, there is minimal infrastructure costs associated with placing the service in a corridor. Infrastructure costs that should be considered however include the ability for the road surface to accommodate bus traffic, stop, seating and passenger shelter infrastructure as well as lighting. These costs are generally already incorporated within typical designs of major roads and arterials. Vehicle costs range for $400,000 for a 12.5 m rigid bus to $600,000 for an 18.5 m articulated bus.

Page 10 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

2.1.3 Operational costs

The operating cost of a standard Australian bus can differ from city to city however; the standard operating cost for a bus is $3.00 to $4.00 Australian dollars per kilometre of operationW. Elements such as maintenance, size of vehicle, age, fuel consumption, drivers wages, management, terrain, type of service (express/all stops) all contribute to the cost of operating a bus.

2.1.4 Value uplift

There is a direct correlation between access to public transport services and increased property values. The value uplift of property adjacent or within walking distance (10 minutes) of conventional bus service is not as measurable as seen with other modes of public transport. In addition, there is little attraction of new development investment due to availability of bus service, especially on low- to mid-frequency routes.

2.1.5 Running ways

Buses are able to operate on almost any street environment in major cities. However, the heavy weights of buses can cause damage to roads which have not been designed to handle the vehicle weight. Buses operate in standard traffic lanes generally in the kerb side lane if on multiple lane roads. Considerations should be made regarding pavement strength and width of lanes when constructing roads to handle buses. Extra attention should be made around bus stops and areas where buses frequently brake as these surfaces, if not designed correctly, can lead to warping and uneven road surfaces.

2.1.6 Corridor reservations

Since buses operate in standard roads, there is little requirement for additional corridor reservations. However, there is a preference for buses to operate in lanes which are around 3.5 m in width. This allows for sufficient room on either side of the vehicle for general manoeuvring.

2.1.7 Stations

In regards to bus stops and stations, bus stops are generally referred to as standard bus stops or bays on the side of the road. Bus stations are generally referred to as locations with larger passenger facilities, where multiple buses operate from and where a transfer between services by passengers is possible. Bus stops are generally spaced between 250–400 m apart and are intended to provide maximum coverage to the community as possible. Bus stops can consist of anything from a single pole or sign on the side of the road to large passenger shelters with real time passenger next bus information, advertising and timetables. Some shelters located in harsh weather conditions can also be enclosed and are air conditioned.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 11

Stirling City Centre Light Rail Feasibility Study - Phase 2

2.1.8 Vehicles

Buses come in various shapes, lengths, widths and heights. From small minivan type vehicle capable of carrying 12 passengers to articulated and double deck buses able to accommodate over 100 passengers. The variation can be significant; however, standard buses in Australia are usually 12.0–12.5 m in length for standard single deck rigids and 18.0 m in length for articulated buses. The standard width of a vehicle in Australia is 2.5 m however; this can vary by plus or minus 10 cm.

2.1.9 Other elements

Power source

Buses generally have diesel internal combustion engines, however, in recent year the advances in technology have lead to new, greener forms of propulsion systems. The alternatives to diesel include Euro standard diesel, Compressed Natural Gas or CNG, Liquefied Natural Gas (LNG), Ultra Low Sulphur Diesel (ULSD), Bio Diesel, Hybrid Diesel Electric, Trolley Bus or Electric, Fuel Cell and Hydrogen. However, some technologies are still in the testing phases while others are mature, costs are still considered to be relatively high in comparison to the standard diesel bus.

Manufacturers

There are numerous manufactures of standard and custom made buses in Australia. The most common manufacturers include, MAN, Scania, Renault, Iveco, Mercedes Benz, Volvo, Mitsubishi and Toyota. In most instances around Australia buses are shipped to Australia in chassis form and then a bus bodies are manufactured here in Australia to suit the client’s requirements. Alternatively fully assembled buses can be purchased directly from the manufacturer.

Passenger loading is conducted through one or more doors on a standard bus. Generally in Australia, street buses will consist of one or two doors for a rigid and two to three doors for an articulated buses. Passenger loading is conducted thought the front drivers door while unloading is generally via the rear doors. Due to fare payment within the vehicle, the loading time can vary widely at a station thereby affecting schedule reliability. However, the introduction of smart card ticketing systems is seeing movement towards loading and unloading from both doors.

Gradients

In most instances buses are generally able to grades of up to 13–15%, with 10% generally applied as practical maximum grade. However, grades of this magnitude generally require more powerful vehicles. CNG buses for example may have the same power to weight ratio as a standard diesel bus, however, the torque curve and power range alter dramatically especially when in demanding situations like hilly terrain. Therefore, short steep sections generally do not affect bus operations, however, longer vertical inclines require operators to consider the types of vehicle specifications required to operate the service.

Page 12 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

Turning radii

The turning radii on most buses would be considered to be good when comparing to other types of mass transit. However, the turning radii of vehicles can vary depending upon manufacture, length of vehicle and the floor height of the bus. Generally buses with less than 12 m in length have the greatest turning ability, followed by articulated buses, standard 12.5 m rigids and then tri-axel 14.5 m length buses. Typical road design standards require that all buses be able to negotiate as 12.5 radius.

2.2 Tram on street

For the purpose of this report, the definition of tram on street refers to a tram or light rail vehicle that operate either in mixed with general traffic or in mixed use lanes with high occupancy vehicles or with other modes of public transport. Terminology in the United States often refers trams as either Trolleys or Street Cars. There is often confusion between the difference between tram and LRT. The report

defines trams as being smaller, slower speed, less prioritised, local serving Photo 2.2 Streetcar (tram) in San Francisco public transport vehicles which operate with parallel bike lane and on-street on steel rails. parking

Street running systems (SRT/trams) are primarily designed as local area circulators, not to serve long haul commuter trips. Along SRT corridors, station spacing is typically within relatively close proximity allowing the LRT to serve as a ‘pedestrian accelerator.’ It can also be described as a ‘horizontal elevator’, similar to a multi-floor department store, with a different purpose or function at each station. A well designed SRT route will show good two- way patronage throughout the day as it supports multiple uses, such as shopping during the day and leisure trips after hours as well as commuter trips during the peaks.

Table 2.2 Typical Street Car/Tram characteristics

Transit type Street transit Typical Maximum Passengers per Hour 13,000 Typical Maximum Frequency (trains/hr) 12 Avg. Passenger Trip Length (km) 1.6 Typical Station Spacing (m) 120–240 Propulsion Energy Diesel, Electric Environmental Considerations Street congestion & minor visual impacts Technological Maturity High

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 13

Stirling City Centre Light Rail Feasibility Study - Phase 2

2.2.1 Patronage capacity Trams come in varying shapes and forms and therefore their capacity varies significantly between different applications. Trams can be single carriages, multiple units or coupled sets ranging from the size of a standard bus to a coupled set of two 40 m length multiple articulated section trams. Therefore the diverse range of tram types available affect the passenger capacity of a tram line. With trams capable of carrying up to 530 passengers, the theoretical peak hour capacity can easily reach 30,000 passengers per direction. However, most tram lines operate with frequencies of 2–5 minutes thus their capacity is between 7,000 and 16,000 respectively3B. However, assuming that a standard tram length of 30–40 m in length is used trams have the capacity for approximately 6,500 passenger per direction per hour with 2 minute headways.

2.2.2 Capital costs The indicative range for capital costs per kilometre of track and infrastructure is approximated to be between $10m and $100m. Examples are provided in Table 2.3 below. However, a significant proportion of the higher cost projects have incurred these cost due to considerable infrastructure requirements such as grade separation, tunnelling, utility relocation and property acquisition. For tram lines which are constructed in new corridors or retrofitted to existing corridors, the cost of construction decreases. The estimate for the new Gold Coast Tram line is between $18m and $22m per kilometres. This general figure usually includes the construction of two tracks, power and signalling infrastructure as well as in most cases depot construction, road widening or alterations and station infrastructure. Tram vehicle costs generally range from $4.5 million for a 30 m vehicle to $6.5 million for a 72 m vehicle.

Table 2.3 Adelaide Street Car/Tram characteristics Approximate capital Daily Tram line Length cost per kilometre passengers Adelaide Victoria Sq – Glenelg 10.8 km 9,500 City Extension 1.4 km $22.1m 11,500 Entertainment Centre Extension (including 2.8 km $35.7m 2,500 bridge widening) 1 Melbourne Box Hill 2.2 km $12.7m Vermont South Extension 3.0 km $10.2m Docklands Drive Extension 1.0 km $7.5m 2 Sydney 7.2 km $15.1m Manchester, UK 37.0 km $23.4m 55,000 Montpellier, France 35.0 km $35.2m 190,000 Portland Streetcar, USA3 Downtown Line 3.84 km $27.6m per kilometre Extension 0.96 km $31.2m per kilometre 5,600

1 http://transporttextbook.com/?p=21 2 http://epress.anu.edu.au/agenda/004/04/4-4-A-4.pdf 3 http://www.lightrailnow.org/facts/fa_por-stc-data-01.htm

Page 14 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

2.2.3 Operation costs

Some studies from around the world have claimed that the operation of trams is cheaper than the operation of buses because of the different power sources. In some cases this may be true however, often the operation costs aren’t fully comparable. The typical Australian average for operational costs in Australia are between $5.00 and $15.00 per vehicle kilometreAH. However, like buses and BRT the actual cost can depend on specialised maintenance, driver wages, management, stop spacing and the power source. There is less emphasis on terrain and size of vehicle than with buses. However, one consideration that is often neglected when comparing BRT to trams is the capacity of the vehicle verses the operating cost. For example to operate a corridor that requires 6,500 passenger per peak hour, a BRT cost per kilometre would be for 60 articulated buses or 30, 40 m length trams. Therefore, the amount of drivers required doubles for the same volume of passengers.

2.2.4 Value uplift

Numerous studies have been conducted whereby the property values around tram stations have increased dramatically. If a high frequency service is provided with a good reputation and public image, property values can increase by as much as 40% as experienced in Portland, USACD. Not only does property values increase but so to can rental returns for office and retail space. Office and retail space rent can be as high as 50% greater around stations than surrounding arterial roadsBM. In most instances trams have increase property values between 7% and 25% with a large proportion of these being at the higher end of the scale.

2.2.5 Running ways

The running ways of trams can be located on numerous different configurations, requiring the “right solution for the right problem” for each line. In other words, there is no single best universal solution. It is also important to recognize that tram systems often evolve over time and preferred running way configurations may change to meet shifting priorities.

Generally trams operate in regular street environments where they are mixed in with regular traffic (similar to conventional bus service). However, trams are not limited to on street running. Some tram systems operate around the world using a combination of running ways. Take Melbourne and Adelaide for example, trams that operate in regular traffic, operate in segregated corridors within the street or like the St Kilda tram or Glenelg tramway, operate within their own corridor off street with grade separated or controlled level crossings. However, trams that operate their majority of service in these separate corridors are referred to as Light Rail Transit rather than trams.

Trams can operate in the centre of the street, along segregated tracks on each kerb side (for two way streets), single kerb side for one way streets; and dual track on a single kerbside. The most common application is centre running due to less conflict with side street traffic, an improved corner radius, resulting in less land take and the ability to combined infrastructure such as stations, signalling (if applicable) and overhead centenary wires at a single location. It generally offers the potential for faster running speed.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 15

Stirling City Centre Light Rail Feasibility Study - Phase 2

Side running tracks allow for the seamless integration of the platform with the foot path, public plazas and retail activities, which can help to activate a street. It also allows for the creation of a narrower street section since platforms can be extended from the footpath in lieu of a small number of parking spaces. It also allows for integration with a kerbside bike lane and on street parking as is the case in San Francisco. However, the split kerb side can be delayed by vehicle turning movements, parking and pick-ups and delivery. The dual track on a single kerbside, such as in Nantes, France, addresses these concerns. It utilizes a dedicated right of way which is seamlessly integrated with the footpath and street activities such as a cappuccino strip.

Trams can also operate in corridors that are shared with other public transport modes. Examples exist in Europe where trams and buses share corridors in busy inner city locations. There are also examples where trams and kerb guided BRT systems share the same corridor.

The most common form of running way bed or trams is either located on slab or concreted track. However, in older cities found in Europe, often the running way are located in paved or pebbled track beds.

2.2.6 Corridor reservations

The right of way required to operate trams is generally between 3.1 m and 4.0 m depending on the situation. This applies to both mixed traffic situations and segregated lane operation.

2.2.7 Stations

There are two station configurations generally used for regular tram stations, central island platforms or side platforms. There are advantages and disadvantages on each platform position, however, there is no overall preference for one or the other. The position is depends on the application to which the tram system is being applied.

Both options are applicable to centre running trams. Centre islands can decrease construction costs but make it more difficult to share the right of way with buses due to opposite door configurations. In addition, the resulting median strip is restricted from any activity. Centre islands are generally raised above, but can be slightly depressed below street level to create virtual platforms as is the case in Reinach Dorf, Germany. Outside platforms allow for integration with buses but result in two outside planting strips that cannot be used for activity. In both cases the, residual strips can be planted to increase visual amenity, but will require maintenance.

Generally where trams are operating in the central median of a road corridor there is a tendency to opt for central island platforms as they offer the benefit of reducing the required width of the corridor as the passenger waiting and loading areas are combined into a single larger platform rather than two smaller kerb platforms. For trams operating in the kerb lane of a roadway, the platforms are generally incorporated within the pedestrian footpaths.

A variation on the side platform is a signalized all-stop street stop such as is used in Reinach Dorf, Germany. The signalized all stops uses the traffic lanes as a platform by stopping all traffic at a signal prior to the station. This allows the pedestrians the opportunity to cross at mid-block locations, but works best in low traffic volume situations.

Page 16 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

The overall length of tram platforms varies from system to system. Most systems design platforms for the largest tram unit that will operate on the system. However, many systems have grown to a point where platforms are required to be extended to enable larger trams to serve them. Generally, tram platforms range from 30 m to 80 m in length. However, it should be noted that if smaller platforms are selected in the initial system, contingencies should be put in place to ensure that the platforms can be extended if future requirements demand them to be.

2.2.8 Vehicles

Street running trams can come in many shapes and forms. Older trams are usually single units and are not wheelchair accessible however, most modern tram are low floor vehicle and are made up of multiple segments. Both old and new trams can be designed to be coupled together to operate a longer consist and increase capacity without increasing frequency. Trams can range in length from 10–12 m for a single section to 55–80 m with multiple sections. Trams in Budapest operate as either single 7 or 9 section vehicles of 40 or 54 m length or as double length sets (two trams coupled together) of 80 m. It has also been demonstrated that up to three tram units can be coupled together and successfully operated as a single 120 m length tram. However, these trams of this length are not practical as station lengths, operation and manoeuvrability also become difficult with vehicles of this size.

2.2.9 Other elements

Power source

The most commonly used form of power supply is from overhead catenaries wires. However, technological improvements to power supply delivery has increased dramatically over the last decade. There are now various forms of power supply that can be used for trams. Some of the alternative power sauces include, ground based power supply (GBPS), which is currently extensively used in Bordeaux, France, where the power is drawn from a third rail that is positioned in the centre of the track. Other alternatives to overhead wires include standard and nickel-hydrogen batteries where power is recharged at points along the route or at stations which can then be used to propel the tram when the overhead power supply is not present. However, these technologies are still developing and the current systems do not allow for high speed operation.

A new potential emerging technology is electromagnetic power supply. This consists of transferring power from the power cables buried beneath the rail to the tram vehicle into magnetic fields which induce an electric current that can be picked up by coils onboard the tram. This process uses the same technology used to power electric toothbrushes. Currently the technology is not in commercial operation however, a test track in Augsburg in Germany is currently under constructionCG.

Manufacturers

There are several manufacturers when it comes to the manufacturing of tram vehicles. They major manufacturers include Bombardier, Siemens, Alstom, Skoda and INEKON.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 17

Stirling City Centre Light Rail Feasibility Study - Phase 2

Trams are designed for cater for large volumes of passenger entering and exiting the vehicles. A typical tram vehicle is designed to have large open areas and multiple doors. For a standard 30 m length tram, it typical to have between three and six doors per tram unit regardless of whether it’s a two or five section tram.

Gradients

Generally most street running trams can operate on gradients of up to 6%. Some trams are able to have gradients of up to 10%; however, these trams do require traction equipment to be installed on all axels to achieve this grade.

Turning radii

Depending on the type of tram selected, tram turning radii range from approximately 15 m to 30 m with most manufacturers falling between 18 m and 25 m. Ideally most tram systems are design for the maximum allowable radius to reduce noise, maintenance and allow for greater speeds around corners.

2.3 Bus Rapid Transit (BRT)

The definition of Bus Rapid Transit (BRT) can vary significantly when comparing to different examples across the world. As shown in Table 2.4 below, BRT can be in various forms, ranging from very simple to complex improvements. However, the elements that identify BRT over standard bus services is the higher quality of service provided to passengers, improved reliability and reduced travel times. BRT systems often provide users with more frequent Photo 2.3 Centre island platform along services over a long period of time and Eugene, Oregon EmX BRT line greater capacity to move larger volumes of passengers. BRT provide fast and efficient public transport service which are able to get passengers to their destinations while providing flexibility in relation to routes, services and capacity. More complex BRT systems have been referred to as ‘light rail on rubber tyres’.

Table 2.4 Range of BRT characteristics

Stations Roadway Service plan Vehicles Systems Simplest ‘Super’ stops, Mixed traffic, Single Buses with Radios, shelter Queue All-stops line Unique Rte. Electronic fare jumpers ID’s, Head boxes Signs Most High platforms, Fully grade- All-stops, Hybrid, Central Control complex P/R, amenities separated On-line Guided, Room, TSP, services Transitway expresses, Specialized CAD, feeder/ Vehicle Smart Cards line-haul Proof of payment Source: S Zimmermann, World Bank

Page 18 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

Table 2.5 Typical BRT characteristics

Transit type Semi-rapid transit Typical Maximum Passengers per Hour 4,000–20,000 Typical Maximum Frequency (veh/hr) 60–80 Avg. Passenger Trip Length (km) 6.0 Typical Station Spacing (m) 400–800 Propulsion Energy Diesel, Euro-diesel, Natural Gas, Biodiesel, Hydrogen, Hybrid Environmental Considerations Noise, visual and minor traffic impacts Technological Maturity High Land Use Integration and Placemaking Low – High Market Value Uplift High (Stations Only) Average Operating Cost per Seat- kilometre <> $1

2.3.1 Patronage capacity

BRT systems vary in size and scale from queue jump lanes on arterial roads to major corridor like in Brisbane Busways in Queensland, Bogota, Columbia and Curitiba in Brazil. The peak hour capacity of a BRT is dependent on the level of infrastructure invested into the system and the frequency in which the service operates. For example the Bogota system claims to be able to handle 67,000 passengers per hourA in the corridor where as the South East Busway in Brisbane is now carrying 20,000 passenger per hour in the peak direction and the Adelaide O-Bahn is capable of handling 7,500 passengersA. However, the Sydney Transit Way T80 BRT route operates at up to 5 to 7.5 minute frequencies in peaks with standard buses with a capacity of approximately 600 passengers per hour. However, the theoretical capacity is closer to 3,000 to 4,000 passengers per hour and the failure to introduce integrated service structure is the main reason why demand is not higher.

2.3.2 Capital costs

Like patronage capacity, and as shown in Table 2.6 below, the capital costs of the BRT network can vary dramatically depending upon the application. If queue jump lanes and signal priority are the only infrastructure built then the construction cost per kilometre can range between a $0.25m to just under $2.0mAF 5. However, should the BRT system feature grade separated crossings, elevated structures or tunnels then the cost can increase up to $158m per kilometre such as the Inner Northern Busway in BrisbaneCH.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 19

Stirling City Centre Light Rail Feasibility Study - Phase 2

Table 2.6 Existing BRT characteristics

Approximate Capital BRT line Length Daily passengers Cost per kilometre Adelaide O-Bahn 12.0 km $8.2m 30,000 Brisbane South East Busway 16.5 km $22.4m 150,000 Northern Busway $104m Inner Northern Busway $158m 55,000 Sydney Transitway 31.0 km $11.2m Curitiba BRT 60.0 km $1.5 Bogota BRT 84.0 km $7.4m – 16.6m 1,600,000 Istanbul BRT 47.0 km $2.0 – $11.0m 850,000

2.3.3 Operational costs

Since most BRT systems use standard buses, the operating costs for these vehicle are very similar to that of regular city buses. However, specialised BRT systems such as the Adelaide O-Bahn or the optically guided busways in Europe have slightly higher the maintenance costs than standard vehicles. These additional costs can however, be offset by the benefits of improved running performance and reduced kilometre costs by increased efficiency. Therefore bus operating costs can vary from $3.00 per kilometre to $19.00 kilometreAE 7.

2.3.4 Value uplift

There have been several studies around the world which have examined the property value increases that BRT systems have on surrounding neighbourhoods. Some BRT systems around the world have recorded significant growth in property values; Brisbane is often cited as a good example of how BRT has influenced values. Properties around the Brisbane BRT have recorded increases of up to 20% when compared to surrounding suburbsBO. However, the Brisbane BRT is a highly prioritised BRT system that has very high service frequencies. Growth on property values along corridors with standard dedicated bus lanes and small scale bus priority are unlikely to attract such high levels of value uplift.

2.3.5 Running ways

BRT does not necessarily mean a bus that operates on a segregated busway. There are numerous examples around the world that demonstrate the various forms of BRT operation.

Some of the best known examples of BRT are located in Bogota, Columbia and Curitiba, Brazil; however, these two systems have been designed to move large volumes of passengers over relatively long distances. They operate completely in their own right of way with mostly at-grade intersections

Page 20 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

The Adelaide O-Bahn, is an example of a guided busway designed to serve long distance commuters. This system operated in its own right of way (specially fitted vehicle are only able to use the guideway), is fully grade separated, has limited stops and a narrow corridor (approximately 8.0-metres). The guideway and limited stop spacing enable high operating and running speeds (average 50–60 km/h operation with maximum 100 km/h operating speeds).

The Brisbane busways are dedicated roadways that allow buses to travel at higher speeds without the restrictions of other general traffic and congestion. The South Eastern busway operates along the South East Freeway in dedicated grade separated lanes. Parts of the South Eastern Busway also operate in dedicated bus only tunnels.

The Sydney Transitway (T-Way for short) has three main lines and a total length of about 60-kilometres. The first line from Liverpool to Parramatta operates in a segregated right of way (in a former road reserve) for about two-thirds of its 30-kilometre length with the remainder featuring dedicated on-road T-way lanes. The Northwest T-way runs besides major arterial roads as well as on street full time bus lanes. The Orange BRT Line in Los Angeles, USA, operates on dedicated lanes however, has at grade crossings which are prioritised for the BRT.

Although, many of the world’s examples involve high levels of capital expenditure, BRT can also refer to simpler, cheaper options such as bus priority lanes, queue jump lanes at congestion hotspots and/or bus priority phasing at intersections. In several systems, dedicated bus lanes are shared with cyclists, motorbikes Photo 2.4 Fully dedicated BRT corridor and/or taxis to make more efficient use of the right of way.

2.3.6 Corridor reservations

The corridor of a BRT system can vary significantly depending upon the type of system adopted. Generally, for dedicated bus lane a 3.0 m to 3.5 m width kerb lane is an adequate width to accommodate most BRT vehicles. For BRT systems operating in central medians or within their own right of way, a 3.5 m to 4.0 m corridor width is recommended. However, as the speed of the BRT increases so too does the required road width. Guided systems can operate at higher speeds in smaller corridors however, require greater amounts of infrastructure. The overall corridor width also will significantly increase at stations unless they are integrated within the roadside pedestrian realm.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 21

Stirling City Centre Light Rail Feasibility Study - Phase 2

2.3.7 Stations

In most BRT systems around the world, the spacing between stations is generally greater than regular bus services. Station spacing is typically between 500–600 m compared with the standard 250–400 m. However, on the long distance commuter based systems, station spacing is similar to LRT or train station spacing of 1 km to 5 km. Stations are also designed to cater for larger amounts of vehicles, greater areas (often sheltered) for passenger waiting and improved security and public transport information (Intelligent Transport Systems technology). Systems like Bogota and Curitiba include fully enclosed platforms with screen doors to replicate a metro style boarding system as well as having ticketed waiting areas, requiring passengers to purchase tickets prior to entering the station but allowing for multiple door loading and unloading.

Stations can be in various shapes and forms ranging from single or double bus bays platforms to multiple zone catering for 10–20 buses. The size of the station depends on the amount of throughput and whether the station is an interchange to other services. Stations can also be designed with overtaking lanes to cater for express buses or skip stop services. Systems in Bogota, Adelaide, Brisbane and Sydney all allow for buses to pass each other at stations, whereas systems in Curitiba do not allow this. The benefit of allow buses to overtake at stations enables overtaking of other services which may require longer passenger loading time, operate on an express or skip stop pattern or have broken-down. However, overtaking lanes significantly increase the land area and right of way required for each station.

Most BRT stations are kerb side loading, meaning that standard buses are able to use both the BRT stations as well as regular road side kerbs bus stops. However, some systems provide centre island platforms in which case the vehicles are specially designed for the BRT only. Some BRT's even have doors on both side of the vehicle for dual side loading or ability to use centre or side running platforms. Centre island platforms provide the ability to merge and reduce the overall required space for passenger loading, thus removing the space required for each station and therefore the overall corridor width is less. However, centre island platforms do provide issues in terms of types of buses that can be used or the operation of the BRT. Istanbul, Turkey operate standard buses on their BRT whilst using centre platforms. However, buses are required to operate on the opposite side of the corridor (or run again the standard traffic flow).

2.3.8 Vehicles

There can be a significant variation between different types of vehicles that use BRT systems. Floor heights are an essential component to any BRT. Most BRT systems around the world utilise low floor or fully low floor (all seating and entry points are at the same height) buses. Older systems such as the Adelaide O-Bahn still use high floor buses with steps however, most of these vehicles are gradually being replaced with low floor models. The South American examples of BRT use fully high floor vehicles with high floor platforms. These buses remain fully accessible with same level boarding with all stations being fully accessible.

Buses can also come in various shapes, lengths and styles. The BRT systems in operation in Australia use standard rigid and articulated vehicles that can penetrate regular streets. Systems in the Los Angeles and Las Vegas in the United States have special stylised vehicles which are sleek, modern and are different to standard city buses, thus giving the impression of a premium service similar to a metro, tram or LRT. Buses can also be of

Page 22 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

varying lengths, again in Australia standard rigid and articulated buses are used (12.5 m and 18.5 m in length respectively), however, BRT systems in Europe and South America use 24 m length double articulated vehicles. These vehicles have a much greater capacity however; their size can restrict their operation to BRT systems only.

Finally there are also several forms of guided vehicle currently in operation on BRT systems. There are three forms of guided bus systems currently around the world. Firstly is the mechanically guided O-Bahn, this system relies on small horizontal wheels running against a solid kerb or guideway. This system has proven to be reliable but maintenance costs of vehicles can be an issue. The benefit to the system is higher speed operation, up to 100 km/h or more and a narrower corridor (8.5 m). The system does have a few drawbacks, firstly there is a significant increase in capital costs and secondly all large crossings must be grade separated. The second guidance system is optical or magnetically guided buses, these buses rely on guidance systems placed both in the bus and within the pavement to guide the vehicle. These systems with are mostly operating in Europe are better suited to lower speed, confined, inner city operation.

2.3.9 Other elements

Power source

Most BRT systems around Australia and the world utilise standard diesel powered buses. The new buses that are introduced into Australian BRT systems generally meet the strict standards that have been applied in Europe on emissions standards (Euro V). However, there are a wide variety of bus options that can be used for BRT vehicles. The second most common form is Compressed Natural Gas which again is widely available in most Australian cities, including Perth. However, greener technologies are being applied to the bus manufacturing industry and more and more alternative options are becoming available. These include hydrogen, fuel cell, electric and bio-diesel. However, the maturity of these alternatives is relatively low and system wide operations are generally uncommon. Another alternative to an onboard fuel supply is to operate electric buses using overhead wiring systems. However, this option increases the capital cost of construction and has similar visual amenity issues to Trams and LRT.

Manufacturers

Since in most instances BRT vehicle are similar if not the same as standard buses, most manufacturers who build standard buses will be able to supply BRT vehicles, this includes some manufacturers who provide stylised vehicles especially for BRT systems.

In most instances on Australian BRT systems passenger are loaded from the front door with passengers exiting any door. Standard buses are equipped with one to three doors and articulated buses equipped with two to four or five doors. Double articulated buses used in Bogota and Curitiba can have as many as seven doors. Since the South American systems have been designed for large passenger volumes, all stations are ticket controlled and require passengers to have tickets. This also enables the vehicles to be loaded and unloaded from any door. This simultaneous boarding and alighting system from any door radically reduces stationary time at stops, thus increasing the passenger throughput per hour.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 23

Stirling City Centre Light Rail Feasibility Study - Phase 2

Gradients

The maximum achievable gradient for BRT buses is similar to that of standard buses. However, long inclines using double articulated buses can cause some issues with increased running times and the physical loading on the engine. It should also be noted that larger volumes of passengers per vehicle also affect the ability for vehicle to accelerate or maintain speed up inclines.

Turning radii

Like standard buses, BRT also have similar turning radii’s. However, double articulated buses may require larger turning circles and clearances than standard rigid and articulated vehicles.

2.4 Light Rail Transit (LRT)

As previously mention there is often confusion between trams and LRT. For the purpose of this report, Light Rail Transit or LRT will be referred to as LRT vehicle with operates either within a dedicated corridor either with the road corridor or within its own reserve. LRT can also refer to elevated or underground systems. For example the fully automated Docklands Light Rail in London or the Dubai Metro is considered to light rail transit rather than metro services. Figure 2.1 Centre median running LRT There can be similarities between the two with outside platforms in definitions and even in some cases a system Portland, Oregon can be both. For example the Glenelg tram line in Adelaide operates as a tram in mixed traffic in Glenelg before becoming an LRT running in a dedicated reserve with level or grade separated crossings, the line then reverts back to a tram when running through part of the city centre, again operating in mixed traffic before returning to LRT half way through the city when it operates in a dedicated centre lane within the road reserve.

Table 2.7 Typical LRT characteristics Transit type Semi-rapid transit Typical Maximum Passengers per Hour 9,000–32,000 Typical Maximum Frequency (trains/hr) 12 Avg. Passenger Trip Length (km) 6.4 Typical Station Spacing (m) 240–5000 Propulsion Energy Diesel, Electric Environmental Considerations Moderate visual impacts and additional right of way Technological Maturity High Land Use Integration and Placemaking Moderate Market Value Uplift High (Stations Only) Average Operating Cost per Seat - Kilometre <$1

Page 24 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

2.4.1 Patronage capacity

Trams and LRT have very similar capacities in terms of both passengers per vehicle and passengers per hour. However, the general difference between trams and LRT vehicles is their size and capacity per individual vehicle is generally greater than that of a street running tram. The added benefit of LRT systems is that often they are segregated from regular traffic but are not limited to operating with traffic. Due to the exclusive right of ways LRT systems are able to operate more frequently with less congestion, improved performance and hence larger passenger throughput. LRT systems can achieve peak hour passenger very similar to trams however, the slight increase in passengers per vehicle and the ability to have longer vehicle stipulates that LRT systems can achieve a throughput of between 30,000 and 40,000 passengers per hour. However, this requires high levels of vehicles per hour and high priority within the corridor.

2.4.2 Construction costs

LRT systems operate in their own segregated corridor, meaning that construction costs can increase or decrease dramatically depending upon the application. For LRT systems operating in their own corridor away from roads and crossings, costs can be as little as $10m per kilometre. However, when LRT systems are placed in road reservations, costs can quickly increase. This is also the case when LRT systems are constructed above or below grade or where grade separated intersections are required. Generally the construction costs of LRT are similar to BRT if they are being proposed in a similar application.

Table 2.8 Existing LRT system characteristics

Approximate capital Daily LRT line Length cost per kilometre passengers Denver, USA South West Corridor4 13.92 km (plus central $43.74m 22,500 line 8.48 km) South East Corridor 30.56 km $34.33m 32,000 East Corridor 36.32 km $25.91m West Corridor $43.74m Gold Line $59.34m Portland, USA 5 Blue Line MAXX 28.8 km 66,300 Red Line MAXX 8.8 km 25,700 Yellow Line MAXX 9.3 km 6 13,600 7 Salk Lake City, USA 16.96 km $37.66m 9,500 8 9 Dubai, United Arab Emirates 75.0 km $50.0m 59,347

4 http://www.cfte.org/success/success_denver.pdf 5 http://www.lightrailnow.org/news/n_newslog2007q2.htm#POR_20070426 6 http://www.lightrailnow.org/news/n_por_2007-07a.htm 7 http://www.fta.dot.gov/regional_offices_9054.html 8 http://dubaimetro.eu/about-dubai-metro 9 http://dubaimetro.eu/featured/3751/dubai-metro-mall-of-the-emirates-station-lifts-the-most-number-of-passengers

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 25

Stirling City Centre Light Rail Feasibility Study - Phase 2

2.4.3 Operation costs

Trams and LRT are very similar in nature and often trams car purchased can be used for either tramway or an LRT system. Therefore, the general costs of a tram would be similar to an LRT.

2.4.4 Value uplift

Like trams LRT systems have a dramatic effect on property values around stations and the corridor. LRT systems are more likely to attract development and increase property values more than conventional bus transit systems. This is due to the fact the LRT systems are seen as more a ‘Mass Transit’ style service rather than a regular public transport service. The segregated right of ways and their ability to overtake congestion provide the public with a perception that they are a more exclusive mode of transport. LRT systems also have higher reliability and are often stylized to improve their image. These characteristic all contribute as a catalyst to development as well as having significant increases to surrounding property values.

2.4.5 Running ways

LRT systems generally operate in their own right of ways either within road corridors or within their own right of way. Like trams and BRT systems there are three locations that LRT tracks can operate. Centre of the road, split to the kerb lanes or double tracks on one side of the corridor. Generally LRT systems, if operating in streets, operate in the central median. This allows for greater operation flexibility and increases the radii's required to manoeuvre the LRT vehicles around corners. Since LRT vehicles are often longer than trams, larger corner radii's of 25 m or more are required.

2.4.6 Corridor reservation

The downside to operating LRT in dedicated lanes in a road corridor is that the additional space required for the LRT system can increase the land required for the corridor. Since LRT operate in a dedicated lane, if a barrier or median is placed between the general street traffic and the LRT tracks, the corridor can increase from the standard 3.5 m per lane to 4.0 or 5.0 m per track. This creates issues when reservations are tight as the corridor often requires 8.0 m to 10.0 m for dual tracks, not including stations.

2.4.7 Stations

Like regular street running trams there are two forms of stations commonly used for LRT systems. They are centre island platforms and side running platforms. The lengths of the platforms are also very similar, with most LRT and tram vehicle being the same overall length. The one difference between the two types of station platforms is their height. Some LRT systems operate with high floor trams which require high floor platforms. However, as most modern cities look for the latest low floor technology, platform heights are generally the same.

Page 26 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

2.4.8 Vehicles

LRT vehicles, like trams come in various shapes and forms. Most LRT vehicles are slightly larger than street running trams. Generally trams are between 2.3 m to 2.65 m in width, depending upon the application LRT vehicles can be as wide as 3.2 m however, most common examples are around 2.65 m. It should also be noted that the wider the LRT vehicle the greater number of passengers they are able to carry however, larger vehicles are harder to manoeuvre within city streets. Some LRT vehicle can be as long as 20.0 m or more, however, like the width of the tram; longer vehicles increase the corner radii required. However, most modern LRT vehicles are made up of multiple sections that can be coupled together. The San Diego LRT often operates their 27.7 m length LRT vehicles in two or three sets. The other main difference between trams and LRT is that LRT vehicle are often designed to achieve higher running speeds. LRT systems can often reach up to 110 km/h where as trams are typically designed for 70 km/h operation.

2.4.9 2.4.9 Other elements

Power source

Like trams, most LRT systems draw power from centenary overhead wires. However, the major difference between trams and LRT is that most LRT vehicles operate using 1.5 kAC rather than the common 750 vDC. This is due to LRT operating at higher speeds in the dedicated corridors. This increased speed and generally higher vehicle weights require a larger power source. Some LRT systems use third rail technology, however, these systems are generally fully grade separated and are not used in regular street running. Alternatively, in recent years technology has allowed for diesel operated LRT routes. This alleviates the requirement for overhead wires or third rail power supplies.

Manufacturers

Like street running trams, the main manufacturers for LRT vehicles are the same. The major companies include Bombardier, Siemens, Alstom, Skoda and INEKON.

LRT vehicles are often designed to serve more the commuter function rather than the local serving street running trams. Therefore, the amount of passengers boarding and alighting at each stop is generally less. Therefore, most LRT vehicle are designed with maximum seating configurations rather than more doors for fastest boarding times. However, LRTs are not limited to this. It is more common to have between two and four doors per 30 m unit rather than up to six doors like street running trams. Depending on the LRT vehicle selected, the ability to load and unload at stations may cause issues at busy stations during peak periods.

Gradients

Generally most LRT vehicles can operate on grades of up to 6%, however, typically LRTs are designed with lower grades for faster operation. Specialised LRT vehicles could reach grades of up to 10% if additional traction equipment is fitted.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 27

Stirling City Centre Light Rail Feasibility Study - Phase 2

Turning radii

Due to LRT vehicle being larger than street running trams, the turning radii for these vehicles is often larger than a trams. Most LRT vehicle requires radii’s of 18.0 m or more however, the standard turning radii for an LRT is 25.0 m.

2.5 Emerging technology – trams on tyres

TransLohr is a relatively new form of public transport technology. TransLohr combines the advantages of rubber tyres, guided vehicles and electric propulsion. There are currently seven cities around the world that have chosen TransLohr as their chosen transport technology, these include Clermont-Ferrand, France, Tianjin, China, Padua, Italy, Venice, Italy, Shanghai, China, Paris, France and Châtillon, France.

The TransLohr technology combines various aspects of standard buses, BRT, Trams and LRT. TransLohr vehicle look and feel like a tram or LRT, this is an advantage as the passenger perception about them would be considered higher than a standard bus or even BRT. Also due to their rubber tyres they are more adaptable for hilly terrain and require less road bed and ground level service infrastructure. The TransLohr vehicles are also able to negotiate a tighter curve which makes them ideal for narrow, confined corridors.

TransLohr is a proprietary technology. Investment in such a system, limits the transit provider to vehicles manufactured solely by TransLohr. This raises concerns as to the long term viability of the system, cost competition among manufacturers and technological maturity.

2.5.1 Patronage capacity

TransLohr has the ability to meet the same capacity requirements as most BRT, tram and LRT corridor. TransLohr vehicles are able to accommodate a maximum of 345 passengers on a 46 m length vehicle. They are able to operate at frequencies similar to trams, therefore, the capacity ranges from 10,000 to 20,000 passengers per hour4.

2.5.2 Capital costs

Unlike trams and LRT, the ground level infrastructure required for a TransLohr is considerably less. Since the vehicles operate using a single central guideway and run on rubber tyres, the cost to construct the ground on which they operate is less. TransLohr running track requires only 20–30 cm of track bed compared to an estimated 70 cm to 1 m for standard LRT and trams. This reduction in track surfacing reduces the cost of construction to less than $10m per kilometre for standard applications to $46m in corridors with high levels of infrastructure like utility relocation and grade separation.

Page 28 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

Table 2.9 Existing TransLohr systems - Costs

Cost Vehicle Capital Length Vehicles Cost AUD Vehicle Euros cost cost Clermont- €150.00m 14.50 km 20 $217.50m STE4 $69.60 $147.90 Ferrand, France

Padua, Italy €90.00m 10.30 km 16 $130.50m STE3 $51.04 $79.46 Venice, Italy €200.00m 20.00 km 20 $290.00m STE4 $69.60 $220.40 Tianjin, China €68.00m 8.00 km 8 $98.60m STE3 $25.52 $73.08

Shanghai, China €80.00m 9.00 km 9 $116.00m STE3 $28.71 $87.29

Operating cost €5.00m 1.00 km 1 $7.25

STE2 €2.00m $2.90m

STE3 €2.20m $3.19m

STE4 €2.40m $3.48m

STE5 €2.60m $3.77m Source: Email by [email protected] (7 June 2010) to Peter Wong (PB)

2.5.3 Operational costs

The overall operating costs of the technology have not yet been documented publically. However, based on the electric propulsion system, it may be assumed that the operating costs for the technology would be similar to that of standard trams or LRT vehicles. However, there may be increased cost per vehicle kilometre as the rubber tyres of the vehicle increase friction between the road surface and the tram thus requiring additional power to accelerate and maintain operating speeds.

2.5.4 Value uplift

As TransLohr technology is relatively new to the industry, detailed market studies on the current three systems that are in current operation have not yet been conducted. However, assuming that the system offers the same operational characteristics as trams and/or LRT, it is reasonable to assume that the value uplift returns would be similar.

2.5.5 Running way

The running way for TransLohr vehicles are similar to that of trams on street. The width of the vehicles are similar and the positioning within the road corridor as also alike. Since TransLohr vehicles are on rubber tyre, there is a slightly wider path in which is needed for tyres. The guiding device is located in the centre of the running way and consists of a single track in which the guidance system is attached to. Like trams and LRT the TransLohr can operate on surfaces such as grassed track beds, concrete slabs or in standard roadways.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 29

Stirling City Centre Light Rail Feasibility Study - Phase 2

Complications have been noted on the existing systems regarding running way wear and tear which has required replacement. Since the rubber tyres are in a fixed location, unlike BRT, the continuous forces placed upon the pavement where the tyres operate have caused degradation to the running way. Also cities such as Padua, Italy have had several issues in regard to daily maintenance of the single track, as the track must be cleaned of debris before operation. In addition, there have been issues with derailments and safety problems for both push bike cyclists and motor cyclists due to the angled groove.

2.5.6 Corridor requirements

Like LRT and trams, the TransLohr system operates in a similar corridor. TransLohr vehicle can operate in either mixed traffic, shared with other modes of public transport or can operate in their own segregated corridor.

2.5.7 Stations

Due to TransLohr vehicles being of similar shape, size and length to regular trams or LRT vehicles, TransLohr stations and platforms are also of similar dimensions. Stations can either be centre or side located and the length can range from 20 m to 50 m. The major differentiation between TransLohr and trams or LRT is that the station height can be as low as 25 cm rather than the standard 30–35 cm for regular trams or LRT vehicles.

2.5.8 Vehicles

TransLohr vehicles come in one standard model which can range in length from a three unit set of 25-metres to a six unit set of 46 m. The capacity of the vehicle range from 170 passengers to 345 total passengers respectively. The greatest benefit to these vehicles is the almost silent operation when running on street, the noise produced is noticeably lower than standard trams and LRT vehicle as the rubber tyres do not have wheel squeal when turning corners.

2.5.9 Other elements

Power source

There are currently one power supply used for TransLohr vehicle, however, obtaining the electricity to operate the vehicles is done in two methods. The first method is the standard catenary system used with trams and LRT system. The second is known as WiPost. This technology basically removes the overhead wires from the system and replaces them with long conducting horizontal poles on the roof of the vehicles. As the tram passes the light pole or power pole, this conducting mechanism draws power from the pole. Each pole is spaced evenly along the corridor with spacing less than the length of the tram to ensure that power is supplied to the vehicle at all times. However, this does become an issue at corners and when crossing roadways or sections where poles cannot be located. The second technology also being used, but not exclusive to TransLohr, is the use of onboard rechargeable battery systems. This allows the vehicle to charge while in normal running then draw upon the stored power for sections where overhead wires are not used.

Page 30 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

Manufacturers

Currently there is only one manufacturer that is producing vehicles for the TransLohr system, this is Lohr industries. Unlike the other modes examined, having a single manufacturer can have issues in terms of flexibility and market competition which can result in increased costs in the longer term. For example the purchase of additional vehicles or the replacement of vehicles at the end of their life for a system may be more expensive than other modes of public transport due to the uniqueness of the technology.

TransLohr vehicles, like trams and LRT have multiple doors for loading and unloading passengers. Current designs for TransLohr vehicle have one door per unit per side. However, as the technology progresses the ability to customise the vehicle to include more or less access points will become available.

Gradients

The benefit of TransLohr vehicle is their ability to climb steep grades. The vehicles are able to climb grades of up to 13% thanks to the rubber tyres. The electric traction motors also enable the vehicle to climb longer distances than standard buses. This makes the TransLohr suitable for hilly terrain.

Turning radii

TransLohr also have a strong advantage over standard tram and LRT vehicles, the TransLohr vehicle is able to negotiate curve radii’s of as little as 10.5 m. This is considerably less than trams and LRT and is comparable to standard and articulated buses. This makes TransLohr adaptable to confined corridors with tight corners. The small radii also facilitate in the reduction of space required for storage and maintenance of TransLohr vehicles. This ultimately results in less land requirements and a reduction in construction costs.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 31

Stirling City Centre Light Rail Feasibility Study - Phase 2

3. Value capture

One of the hurdles to implementation of quality transit corridors is the large capital investment required by transit infrastructure owners, service providers and local authorities to provide the necessary service and amenities. Revenue from transit services rarely covers capital cost of infrastructure, even over a very long pay-back period, so the majority of funding is normally provided from general tax revenue.

Evidence from built projects clearly illustrates the effect of permanent transit infrastructure on land value is considerable. Value gained by landowners and businesses near transit services can be captured and used to fund transit improvements As a result, governments around the globe recognize that an address near a transit station is a good one. Properties within 400 m of a transit station enjoy improvements in land values of over 50% in comparison to locations away from transit(i) In general, the more accessible a property is to transit, the higher its value.

Business income and revenue and rental return within these precincts is also higher than for those away from transit. There are many examples where this increase in ‘value’ is captured and used to fund transit projects. One example is the Hong Kong rail transit system. It pays all of its costs with value captured from development in station areas.

The potential uplift as well as the catalytic effect of permanent transit infrastructure is discussed by mode in Section 2 above. The purpose of this chapter is to summarize the mechanisms available for value capture, including:

 Joint Development

 Benefited Area Charges

 Tax Increment Financing

 Revenue Sharing

 Developer Contributions

 Parking Surcharges

 Transit User Fees

 Density Bonuses.

Each applies to different sectors in the community and for different components of the infrastructure or service, and each rewards different parties for their part in providing transit. These mechanisms work best when applied together as part of a coordinated strategy.

A number of the potential value capture mechanisms and their application in Stirling are described below. Some of these mechanisms would require changes to state legislation to be used in Stirling, and this is noted. For this purpose, transit service providers and transit infrastructure owners are assumed to be government.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 33

Stirling City Centre Light Rail Feasibility Study - Phase 2

3.1 Joint development

Joint development is the simplest and most easily applicable form of value capture. It is development of land within the transit precinct by or partly by the government. The value of development increase is captured directly by the government through its involvement in development. Joint development has many variations, and can involve the government as land owner, as an investor in development, or both. Usually the government will partner with a commercial developer. The value of joint development is two-fold in that it provides real estate returns, but it also facilitates transit ridership (and therefore revenue) by stimulating development in transit precincts.

As a land owner, the government is able to participate in joint development for the station site and surrounds (including volumetric space above and below station infrastructure) as well as any unrelated government land assets within the transit precincts. Government can also actively acquire land within transit zones for this purpose. Acquisition of land prior to realisation of transit infrastructure affords the greatest financial gain. Extra gain can also be achieved through the acquisition of small, fragmented parcels over a long period of time. These can be used to facilitate partnerships with developers at a later time. Sites used as construction staging areas prime sites for later joint development.

As an investor, the government can contribute funds to development to assist developers to realise a site’s potential, and to reduce their risk. In return, the government is able to benefit from sale profits. Joint development requires significant up front costs by the government and has financial risks for both the developer and government.

3.2 Benefitted areas charges

Areas of existing development around new or improved transit facilities that will benefit from new transit infrastructure and services can be subject to benefitted area charges. Benefitted areas are areas of existing development around stations with new or improved services that benefit from increased value and revenue as a result of those services.

The charges apply to an area along a transit line or around a station in which property owners agree to pay annual assessments over a period of time in exchange for public physical improvements. A key aspect of this tool is that they typically require a majority vote of the affected property owners as a condition of adoption. The amount of the assessment must directly relate to the cost of the improvement and the benefit gained by the property owner.

The districts offer the benefit of capturing the financial return with low or risk. In order for the jurisdiction to get the funds up front to pay for the improvements, municipal bonds, typically 30-years, are issued and the assessments pay back the bonds. The Los Angeles County MTA is one example of a transit agency using a benefit assessment district raised $130 million, or 9% of the funds for the Segment 1 of the Metro Red (A) Line.

This is a mechanism available only to local jurisdictions. This mechanism would be useful when a local authority wanted to inject investment into existing area with existing or new transit. However, if a local authority wished to provide financial support the state government to provide major new infrastructure, this mechanism could be used.

Page 34 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

In the United States, benefitted areas charges are the most common value capture strategy used to fund the costs of new transit, especially for light rail. When used to fund transit facilities, a tiered assessment rate, based on proximity to the line, is commonly used. Owner occupied residential properties are typically exempt from the charges to avoid burdensome taxes and to reduce the number of votes needed to approve the district.

The tool is easier to implement for smaller geographic areas and within single jurisdictions due to this support requirement. Areas with a significant number of property owners who plan to develop or redevelop their parcels are likely areas for these charges. Such owners are typically more motivated to participate in infrastructure investments that will enhance the value of their property or make development financially possible where it would not be (e.g. due to parking requirements that can be waived due to the new transit line).

The Brisbane City Council’s Suburban Centres Improvement Program (SCIP) uses benefitted areas charging as a mechanism to fund local streetscape and public works improvements in suburban centres.

3.3 Tax Increment Financing

Tax Increment Financing (TIF) involves the fixing of the assessed rates value in a defined area that is given to general rates revenue for a period (e.g. ten years) and using any additional revenue from increased property values to fund transit infrastructure or local improvements. Where the local authority cannot pre-fund the improvements themselves, they are sometimes able to borrow against the future assessed rates value. It is particularly useful in areas that suffer from a lack of investment due to high development risks and low returns on investment, or for projects that require place-making infrastructure (streetscapes, sidewalks, parks). These areas require short term investments that have long term return.

Unlike benefitted areas charges, the principal purpose of a TIF district is to encourage new development. Consequently, the goals of most TIF projects are typically broader than a single transit investment. The districts can also play a key role in generating ridership through pedestrian and streetscape improvements, and investments that increase the viability of transit oriented development around stations.

The districts capture the total value of growth in property taxes in a designated area. Thus, new development has a much greater impact on tax revenue than growth in value of existing properties.

This is a mechanism available only to local authorities. However, if a local authority wished to provide financial support the government to provide major new infrastructure, this mechanism could also be used.

Case Study – Transit Revitalization Investment Districts

In 2005, the State of Pennsylvania in the United States authorized the use of Tiffs in Transit Revitalization Investment Districts (TRIDS) around transit stations. The purpose of the tool is to promote TOD and transportation improvements near transit stops. The legislation allows transit agencies to work cooperatively with local jurisdictions to create Tiffs and to share in the value capture. A transit agency may acquire property within the district for real estate development or joint development.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 35

Stirling City Centre Light Rail Feasibility Study - Phase 2

3.4 Revenue sharing

Transit service providers and transit facility owners can enter into revenue sharing arrangements with developers and land owners to help fund transit infrastructure and service improvements. Examples include:

Facility connection/station interface fee – A fee charged by the transit facility owner to the tenant or owner of a retail tenancy or shopping centre for connection of the tenancy or shopping centre with the station concourse or other facilities where increased patronage and therefore financial gain can be demonstrated. The fee can be based on rental return.

New development contributions – Infrastructure contributions charged by local authorities for new development is the established form of cost sharing for local infrastructure in Stirling.

Transit impact fee – a fee charged to new development that has a high percentage of transit users (e.g. office building) where increased use of facilities can be demonstrated.

3.5 User fees

Car parking surcharge – A surcharge is applied to paid car parking charges across the transit precinct. Revenue is directed into transit or other related improvements.

Transit rider fee – A fee charged by the transit service provider or transit facility owner to the transit rider on top of the trip fare for use of specific infrastructure or services that improve the rider experience. This is similar to tolling a road improvement. This could be implemented through selective movement of zoning boundaries, or general increase in fares for use of improved services (e.g. use of a busway v on-street bus). Alternative routes and services might need to be considered to ensure that the service remains equitable. Service fees can also be charged between transit service providers and transit facility owners, if these parties are different.

3.6 Other Innovations

Other innovative strategies that could be explored include the following:

 Grant density bonuses to developers who contribute to rail implementation.

 Establish public-private consortium responsible for both rail infrastructure and station district real estate.

 Government assets - redevelopment of government assets is an excellent opportunity to catalyse development at the same time as capitalising the value of these assets. Possible assets to be disposed of or used as demonstration projects include the state owned land reserved for the Stephenson Highway.

 Land acquisition – the Government has powers under various pieces of legislation to acquire land for its community purposes. Some of these powers (e.g. for transport infrastructure) could be used to acquire land at stations to ensure that these are integrated with above ground development. Legislation could also be amended to allow the Government to acquire land for community development.

Page 36 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

 Street/public space improvements – the Stirling Council can invest money in public works to improve the pedestrian environment in public places around the stations. This will contribute greatly to the attractiveness and usability of these areas and support the pedestrian and transit oriented environment.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 37

Stirling City Centre Light Rail Feasibility Study - Phase 2

4. Refined patronage forecasts – TODTrips model

4.1 Background

As part of the Stirling City Centre (SCC) - Light Rail Feasibility – Phase 2 study, PB conducted a desktop assessment of five alternative modal and operating environments options that could be available to serve the SCC. The range of options starts from bus on street, street car (tram on street), bus and light rail in exclusive lanes with priority (BRT and LRT) and light rail on single side (LRT sided).

This chapter summarizes the development of the public transport model using PB’s TODTrips package to estimate number of passenger trips for the 2031 proposed land use plan and ridership of the five alternative modes. Five alternative modal scenarios were analysed including:

 Base case with 2031 bus option

 Street Car

 LRT

 BRT

 LRT (Single sided).

Each scenario is further described in Section 4.3 below. Additional detail into the modelling assumptions and methodology is provided in the TODTrips Working Paper provided in Appendix D.

4.2 TODTrips model

The TODTrips model has been developed for the study area in consultation with the Stirling Alliance team. Input from other stakeholders has also been facilitated through a series of transport modelling meetings and workshops. The model has been developed to test a range of public transport scenarios for the Stirling City Centre. The main principle of the TODTrips package is the combination of detailed mode choice modelling with assumptions about trip generation, distribution and car travel attributes based on the Department of Planning’s STEM strategic transport model. This approach is designed to allow the rapid development and testing the relative performance of a range of scenarios based on future assumptions regarding land use and transport.

The development of assumptions regarding future land use and transport is an important part of the model development process. It is intended to inform the specification of assumptions regarding the 2031 study area.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 39

Stirling City Centre Light Rail Feasibility Study - Phase 2

The following outlines the broad capabilities and limitations of the TODTrips model:

Capabilities

 detailed modelling of land use patterns, including the distribution of population and employment inside the study area

 detailed modelling of transport networks inside the study area including walk access, public transport lines and services bus, street car LRT, BRT routes and services, and rail stations and services

 sophisticated generalised cost mode choice model for four main modes of travel – car, public transport, cycle and walk

 modelling of daily trips for all purposes that start and/or end in the study area.

Limitations

 coarse representation of zones outside the study area – modelling of links to key origins and destinations

 no modelling of detail road and traffic network – assumptions are made about car travel attributes

 no modelling of trips that start and end outside the study area – that is, no modelling of through trips.

Key features associated with the TODTrips modelling platform applicable to the Light Rail Feasibility – Phase 2 study include:

 modelling of land use according to zoning types and floor space ratios

 modelling of higher density land use around selected transport nodes

 detailed modelling of bus and rail services together with consideration of alternative transit options including Street Car, Light Rail (LRT), BRT and LRT on single side

 modelling of public transport travel to key regional zones around study area

 estimation of daily trip patterns

 mode choice modelling based on generalised cost for car, public transport, cycling and walk trips.

The primary focus of TODTrips is the rapid development and testing of a range of scenarios related to land use and transport planning.

The TODTrips model focused on the average daily person trips for all trip purposes for year 2031 and incorporates the following main modes of travel:

 car – driver and passenger combined

 walk and cycling combined

 rail (access via Glendalough and Stirling Stations within study area)

Page 40 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

 bus (include local and regional bus services that pass through the study area), and

 alternative transit options including bus, street car, LRT, BRT and LRT on single side.

4.3 Assessment result of five transit modal scenarios

Five scenarios were setup in TODTrips to represent five alternative modal options and operating environments that could be considered to serve the SCC in 2031. In general, the broad operating environment for each scenario was set up to maintain appropriate existing public transport services with the addition of a new mode with a specified level of service.

Existing public transport services within the study area are described in Table 4.1 and Figure 4.1. The Northern rail service has Stirling and Glendalough stations as the key rail access points within study area. The speed and frequency of the Northern rail service are based on Transperth time table. The operating characteristics of bus services (including 98/99, 413 and 400) such as speed, number of stops and frequency of services were made available to the TODTrips study team by the PTA.

The section of the Northern Railway Line to Joondalup between Stirling and Glendalough stations, local route 413 bus service, Circle Bus route 98/99 and route 400 were included in every scenario.

Table 4.1 Existing public transports operating environment for Stirling Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 4 3 413 Local bus 19.5 9 4 4 400 Regional bus 24.2 4 4

Figure 4.1 Stop locations of existing public transport services

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 41

Stirling City Centre Light Rail Feasibility Study - Phase 2

These existing public transport services were included in every scenario and operating environment which will be described in the following sections.

4.3.1 Scenario 1 – Base case (Bus)

Table 4.2 Scenario base (Bus’s) operating environment

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 4 3 413 Local bus 19.5 9 4 4 400 Regional bus 24.2 4 4 5 Base Local bus 15.0 13 4

Apart from the existing public transport services (Joondalup rail, 98/99, 413 and 400 bus services), new service is highlighted in Table 4.2 as Service number 5 and labelled as Base. In this scenario, a local bus service is tested with 13 stops (see Figure 4.2), average speed of 15 kph and run every 15 minutes. This speed is a result of the absence of a dedicated running way for busses in this scenario as well as the relatively high number of stops.

Figure 4.2 Stop locations of new local bus service in Scenario 1

Page 42 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

4.3.2 Scenario 2 – Street Car/Tram

Apart from the existing public transport services (Joondalup rail, 98/99, 413 and 400 bus services), a new service is highlighted in Table 4.3 as Service number 5 and labelled as S2. In this scenario, a street car (or tram) is tested with 19 stops (see Figure 4.3), average speed of 15 kph and run every 5 minutes. It should be noted that the operating speed of service 400 is reduced from 24.2 kph down to 15 kph mainly due to safety as both the street car and service 400 could share the road space on the Scarborough Beach Road and general traffic including bus service 400 would have to give way to passengers alighting and boarding at tram stops.

Table 4.3 Scenario S2 (Street Car/Trams) operating environment

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 4 3 413 Local bus 19.5 9 4 4 400 Regional bus 15.0 4 4 5 S2 Streetcar 15.0 19 12

Figure 4.3 Stop locations of Street Car service in Scenario 2

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 43

Stirling City Centre Light Rail Feasibility Study - Phase 2

4.3.3 Scenario 3A – LRT

Apart from the existing public transport services (Joondalup rail, 98/99, 413 and 400 bus services), a new service is highlighted in Table 4.4 as Service number 5 and labelled as S3A. In this scenario, a dedicated LRT is tested with 11 stops (see Figure 4.4), average speed of 20 kph and run every 5 minutes. It should be noted that the operating speed of service 400 is reduced from 24.2 kph down to 20 kph mainly due to safety as both the LRT and service 400 would share the road space on the Scarborough Beach Road. However, the planned speed for the bus service 400 is still higher than in Scenario S2 with Street Car as passenger alighting and boarding at LRT stops will be more protected with LRT.

Table 4.4 Scenario 3A (LRT’s) operating environment

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 3 3 413 Local bus 19.5 9 4 4 400 Regional bus 20.0 4 9 5 S3A LRT 20.0 11 12

Figure 4.4 Stop locations of LRT service in Scenario 3A

Page 44 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

4.3.4 Scenario 3B – BRT

The operating environment of this scenario is identical to Scenario 3A in terms of existing public transport services and the alignment and associated stopping patterns of the new mode (see Figure 4.2). The only difference is that the new mode is changed from LRT to BRT and the frequency of service is reduced from 12 to 4 services per hour.

Table 4.5 Scenario 3B (BRT’s) operating environment

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 3 3 413 Local bus 19.5 9 4 4 400 Regional bus 20.0 4 9 5 S3B BRT 20.0 11 4

Figure 4.5 Stop locations of BRT service in Scenario 3B

4.3.5 Scenario 4 – LRT (Single Sided)

This scenario was constructed as a variation to Scenario S3A which is also a LRT option. However, the LRT proposed in this scenario is only operating on the southern side of Scarborough Beach Road (western side on Stephenson Blvd). This mode of operation might create some distance constraints for residents on the northern side of Scarborough Beach Road as they might need to walk extra distance to pedestrian crossing to be able to cross Scarborough Beach Road. This extra walking distance (currently assumed to be 30 metres) was added to the utility function for those zones affected by this scenario.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 45

Stirling City Centre Light Rail Feasibility Study - Phase 2

Table 4.6 Scenario 4 (LRT to one side) operating environment

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 3 3 413 Local bus 19.5 9 4 4 400 Regional bus 15.0 4 9 5 S4 LRT Side 17.5 16 12

Figure 4.6 Stop locations of LRT (to one side) service in Scenario 4

The assessment of five scenarios was implemented with two car use scenarios: high and low car use for internal trips and high car use for external trips. Results of TODTrips model runs output for all scenarios are presented and discussed in the following sections.

4.4 Internal trips – mode share and ridership estimates

This section presents mode share and ridership estimates for all five scenarios to accommodate approximately 92,000 daily person trips estimated for internal travel movements (I-I). Tables 4.7 and 4.8 present the mode share results based on the high and low car scenarios. Main findings are as follows:

 Walk and cycle mode share is quite consistent and stable across different scenarios with average of 26.8% and 27.9% share in low and high car scenarios, respectively.

 Car, rail and bus had highest share in base scenario among all five scenarios.

Page 46 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

 With the introduction of alternative transit modes including Street car, LRT, BRT and LRT single sided, mode choice pattern was redistributed where alternative modes gain an average share of 20%.

 These gains came from the drops in car share by around 7%, walk and cycle mode share by around 2%, rail share by around 3.5% and bus share by 3 to 8%.

Table 4.7 Low car use – Mode share for internal trips

by Walk by Scenario by Car by Rail by LRT by BRT by Bus & Cycle Street Car Base 47.3% 28.0% 8.3% 0.0% 0.0% 0.0% 16.4% Street car 40.6% 26.4% 3.2% 24.5% 0.0% 0.0% 5.2% LRT 40.8% 26.5% 3.1% 0.0% 22.3% 0.0% 7.3% BRT 41.1% 26.6% 3.6% 0.0% 0.0% 20.5% 8.2% LRT (single sided) 40.5% 26.4% 3.1% 0.0% 24.0% 0.0% 5.9%

Table 4.8 High car use scenario – Mode share of internal trips

by Walk by Scenario by Car by Rail by LRT by BRT by Bus & Cycle Street Car Base 54.9% 29.6% 6.6% 0.0% 0.0% 0.0% 8.9% Street car 45.3% 27.4% 2.8% 20.4% 0.0% 0.0% 4.2% LRT 45.6% 27.5% 2.7% 0.0% 18.6% 0.0% 5.7% BRT 46.2% 27.7% 3.0% 0.0% 0.0% 16.7% 6.4% LRT (single sided) 45.3% 27.5% 2.7% 0.0% 19.9% 0.0% 4.7%

Tables 4.9 and 4.10 provide a summary of ridership share of 92,000 daily person trips among different transport modes for low and high car use. Main findings are as follows:

 Apart from highest ridership values in car, rail and bus, alternative transport modes results in base scenario, alternative transport modes (street car, LRT, BRT and LRT single sided) ridership values to accommodate 92,000 internal person trips are around 20,000 person trips per day in low car use and around 18,000 person trips per day in high car use scenarios.

Table 4.9 Low car use scenario – ridership share among different transport modes for internal trips

by Walk by Scenario by Car by Rail by LRT by BRT by Bus & Cycle Street Car Base 43575 25756 7626 0 0 0 15095 Street car 37331 24342 2965 22588 0 0 4826 LRT 37543 24399 2847 0 20563 0 6699 BRT 37816 24457 3300 0 0 18891 7588 LRT (single sided) 37327 24345 2874 0 22101 0 5406

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 47

Stirling City Centre Light Rail Feasibility Study - Phase 2

Table 4.10 High car use scenario – ridership share among different transport modes for internal trips

by Walk by Scenario by Car by Rail by LRT by BRT by Bus & Cycle Street Car Base 50575 27233 6096 0 0 0 8148 Street car 41667 25260 2546 18744 0 0 3835 LRT 41976 25343 2442 0 17080 0 5212 BRT 42493 25453 2807 0 0 15408 5892 LRT (single sided) 41695 25271 2454 0 18277 0 4355

4.5 External trips – mode share and ridership estimates

This section presents mode share and ridership estimates for all five scenarios to accommodate some 210,000 daily person trips estimated for external travel movements (I-E and E-I movements). Table 4.11 presents the result for the distribution of regional access and egress by external trips into and out of the study area. Main findings are as follows:

Table 4.11 Distribution of regional access and egress by external trips into and out of study area (I-E and E-I movements)

Daily Daily Daily Total pc by pc by trips by trips by pc by Scenario trips by daily regional regional regional regional car Car trips rail bus rail bus Base 135900 46231 26979 209110 65.0% 22.1% 12.9% Street car 131015 48611 29483 209110 62.7% 23.2% 14.1% LRT 129549 47731 31831 209110 62.0% 22.8% 15.2% BRT 130134 47312 31664 209110 62.2% 22.6% 15.1% LRT (single sided) 129623 47630 31856 209110 62.0% 22.8% 15.2%

Among the 210,000 daily person trips estimated for external travel movements (I-E and E-I movements), an average of 130,000 car trips had direct access between external and internal zones. While 130,000 daily person trips of car mode to and from external zones will become internal travel component within the study area, the 80,000 (=210,000-130000) daily person trips by regional rail and or regional bus will be connected to the local transit network within the study area. Tables 4.12 and 4.13 present the result for the modal split and ridership share by local transit modes from these approximate 80,000 daily person trips in connecting to the local transit services. Main findings are as follows:

 In Base scenario, rail and bus mode shares are 27.9% and 7.1%, respectively.

 In other alternative mode scenarios including street car, LRT, BRT and LRT single sided, rail and bus modes drop their share values down to 15.3% and 5.7%, respectively. Alternative transport mode shares gains from significant drop in rail share (12.6% reduction), car share (3% reduction) and bus share (1.4% reduction) with 16.3% mode share for street car and LRT and 12.7% mode share for BRT.

Page 48 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

 Ridership estimates for alternative transport modes ranges from average of 26,000 daily person trips (BRT) to 34,000 daily person trips (street car, LRT and LRT single sided).

Table 4.12 Modal split of external trips using local transit services

by Scenario by Car by Rail by LRT by BRT by Bus Street Car Base 65.0% 27.9% 0.0% 0.0% 0.0% 7.1% Street car 62.7% 15.3% 16.3% 0.0% 5.7% LRT 62.0% 16.3% 0.0% 15.5% 0.0% 6.2% BRT 62.2% 18.0% 0.0% 12.7% 7.0% LRT (single sided) 62.0% 15.8% 0.0% 16.3% 0.0% 5.8%

Table 4.13 Ridership estimates of external trips (I-E and E-I movements) using transit services

by Scenario by Car by Rail by LRT by BRT by Bus Street Car Base 135900 58280 0 0 0 14931 Street car 131015 31901 34183 0 0 12011 LRT 129549 34097 0 32470 0 12995 BRT 130134 37668 0 0 26635 14674 LRT (single sided) 129623 33081 0 34176 0 12230

4.6 Combined internal and external trips - mode share and ridership estimates

This section presents mode share and ridership estimates for all five scenarios to accommodate a total of around 300,000 daily person trips estimated for all travel movements (including internal and external trips). Tables 4.14 and 4.15 present the mode share results based on the high and low car scenarios. Main findings are as follows:

 All for alternative transport modes including street car, LRT, BRT and LRT single sided have gained a high mode share values (from 14% with BRT to 19% with Street Car) in comparing to the bus option used in the Base scenario.

 The 5% difference between BRT and the street car and LRT is mainly due to the frequency of service of BRT is effectively 7.5 minutes (within the dedicated running way) versus 5 minutes. This shows that the modelling results are sensitive to frequency, as a key driver of demand and which underlines the fundamental drivers of success.

 Street car mode gains highest mode share is mainly due to its service coverage with 19 stops in comparing to 13 stops in LRT scenario.

 In terms of ridership estimates, all alternative modes scenarios are comparable and their values are in the range of 45,000 to 55,000 person trips per day.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 49

Stirling City Centre Light Rail Feasibility Study - Phase 2

Table 4.14 Low car use – mode share for combined internal and external trips

by Walk by Scenario by Car by Rail by LRT by BRT by Bus & Cycle Street Car Base 59.6% 8.6% 21.9% 0.0% 0.0% 0.0% 10.0% Street car 55.9% 8.1% 11.6% 18.9% 0.0% 0.0% 5.6% LRT 55.5% 8.1% 12.3% 0.0% 17.6% 0.0% 6.5% BRT 55.8% 8.1% 13.6% 0.0% 0.0% 15.1% 7.4% LRT (single sided) 55.4% 8.1% 11.9% 0.0% 18.7% 0.0% 5.9%

Table 4.15 High car use – mode share for combined internal and external trips

by Walk by Scenario by Car by Rail by LRT by BRT by Bus & Cycle Street Car Base 61.9% 9.0% 21.4% 0.0% 0.0% 0.0% 7.7% Street car 57.3% 8.4% 11.4% 17.6% 0.0% 0.0% 5.3% LRT 57.0% 8.4% 12.1% 0.0% 16.5% 0.0% 6.0% BRT 57.3% 8.5% 13.4% 0.0% 0.0% 14.0% 6.8% LRT (single sided) 56.9% 8.4% 11.8% 0.0% 17.4% 0.0% 5.5%

Table 4.16 Low car use – ridership estimates for combined internal and external trips

by Walk by Scenario by Car by Rail by LRT by BRT by Bus Total & Cycle Street Car Base 179474 25756 65905 0 0 0 30026 301162 Street car 168346 24342 34866 56771 0 0 16836 301162 LRT 167092 24399 36944 0 53033 0 19694 301162 BRT 167949 24457 40968 0 0 45526 22261 301162 LRT (single sided) 166950 24345 35955 0 56277 0 17635 301162

Table 4.17 High car use – ridership estimates for combined internal and external trips

by Walk by Scenario by Car & Cycle by Rail Street Car by LRT by BRT by Bus Total Base 186474 27233 64375 0 0 0 23079 301162 Street car 172682 25260 34447 52927 0 0 15845 301162 LRT 171524 25343 36539 0 49550 0 18206 301162 BRT 172627 25453 40475 0 0 42043 20565 301162 LRT (single sided) 171317 25271 35536 0 52453 0 16585 301162

Page 50 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

5. Summary

In Phase 1, the base case analysis showed an estimated light rail patronage of approximately 27,000 trips on an average weekday. To place these findings in context, comparison was made with light rail systems introduced in recent years in the United States). Comparison with these figures indicates that the Stirling light rail system is definitely ‘in the ballpark’.

The Phase 2 study reinforces the findings that a high quality transit system, such as a LRT or street running tram is viable if supporting land use and transport policies are in place.

The following conclusions and observations are made in relation to the results of the modelling:

 There is a potentially strong market for a high quality transit system to provide for travel within the study area and to facilitate the use of public transport for access to the area from other parts of the metropolitan area. In particular, there is strong potential for the operation of an effective internal transit system in the Stirling Centre. In other words, the transit system will function well as a ‘pedestrian accelerator’ or for relatively short trips amongst origins and destinations within the study area.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 51

Stirling City Centre Light Rail Feasibility Study - Phase 2

 The transit system has a strong role to play in minimising the use of private motor vehicles for movement within the centre and minimising the demand for parking.

 It is essential that supportive land use framework be in place prior to implementation to allow the catalytic effects to be produced effectively.

Page 52 10-0477-02-2106689A PARSONS BRINCKERHOFF

Stirling City Centre Light Rail Feasibility Study - Phase 2

However, verification of this observation will require additional refinement based on the actual proposed operating plan and concept design in Phase 3.

PARSONS BRINCKERHOFF 10-0477-02-2106689A Page 53

Appendix A

Mode comparison summary

Appendix A Mode comparison summary

Mode comparison summary Bus on street Bus Rapid Transit (BRT) Trams on street Light Rail Transit (dedicated) TransLohr

Transit mode use in other cities Currently operating in Perth in every major city O-Bahn, Adelaide Adelaide Adelaide Clermont-Ferrand, France around the world Busway, Brisbane Melbourne Melbourne Tianjin, China LPT, Sydney Sydney Paris, France Padua, Italy Curitiba, Brazil London, UK Barcelona, Spain Venice, Italy Bogotá, Columbia Manchester, UK Buenos Aires, Argentina Shanghai, China Istanbul, Turkey Montpellier, France Boston, USA Paris, France Leeds, UK Strasbourg, France Docklands Light Rail, UK Châtillon, France Nancy, France Stockholm, Sweden Boston, USA Ottawa, Canada Zurich, Switzerland Calgary, USA Pittsburgh East Busway, USA Portland Streetcar, USA Denver, USA Boston, USA Boudreaux, France KCRC Light rail, Hong Kong Los Angeles, USA Portland, USA Kuala Lumpur, Malaysia Porto Alegre, Brazil Baltimore, USA Quito, Ecuador Virginia, USA Sao Paolo, Brazil Los Angeles, USA Kunming Busways, China Pittsburgh, USA Seattle, USA Salt Lake City, USA Charlotte, USA Dubai, UAE Capacity – Vehicle type Standard Rigid 12.5 m length Standard Rigid 12.5 m length 3 Section Vehicle 30 m length3 2 Section Vehicle 20 m length6 3 Section Vehicle 25 m length . 45 seated . 45 seated . 64 seated . 30 seated . 60 seated . 30 standees . 30 standees . 115 standees . 127 standees . 110 standees . 75 total . 75 total . 179 total . 157 total . 170 total

Tri-Axel Rigid 14.5 m length Tri-Axel Rigid 14.5 m length 5 Section Vehicle 40 m length3 3 Section Vehicle 30 m length7 4 Section Vehicle 32 m length . 55 seated . 55 seated . 72 seated . 68 seated . 80 seated . 35 standees . 35 standees . 143 standees . 168 standees . 150 standees . 90 total . 90 total . 215 total . 236 total . 230 total CK 6 Section Vehicle 74 m length 3B Articulated 18.0 m length Articulated 18.0 m length 6 Section Vehicle 54 m length . 180 seated 5 Section Vehicle 39 m length . 65 seated . 65 seated . 58 seated . 420 standees . 100 seated . 55 standees . 55 standees . 296 standees . 600 total . 190 standees . 110 total . 110 total . 352 total . 290 total

3B Double Articulated 24.0 m length 9 Section Vehicle 72 m length 6 Section Vehicle 46 m length . 80 seated . 90 seated . 120 seated . 100 standees . 440 standees . 225 standees . 180 total2 . 530 total . 345 total4

PARSONS BRINCKERHOFF 10-0477-01-2106689A Page A.1

Appendix A Mode comparison summary

Bus on street Bus Rapid Transit (BRT) Trams on street Light Rail Transit (dedicated) TransLohr Peak hour capacity – Passengers per direction per hour A A Indicative range: 1,000 – 20,000 per hour (BRT – exclusive ROW) Indicative range: 4,000 – 25,000 (segregated - exclusive ROW) Indicative range: A AH A AH < 3,000 (bus on street) 2,000-10,000 (US, BRT/bus lanes) 4,000 – 12,000 (tram) 7,000-18,000 (US, exclusive ROW) 4,000 – 12,000 (TransLohr) AH E 1,000-3,000 (US, bus: mixed traffic) 10,000 – 12,000 (U.S – bus. Small because don’t 3,000-6,000 (estimated for Gold Coast) AM take small headways into account) 3,000-14,000 (US, on-street ROW)AH. AT 1,500 (TVM, Paris) 2,160 (Adelaide) AT 2,000 (Route 5 - Hamburg) 4,000 (T2, Paris)AT AE 2,500 (London - bus) 6,000 (Yellow Line, Porto)AT AT 2,800 (L 12, Utrecht) 6,000 (Strasbourg)AA AT 3,300 (Teor, Roeun, Paris) 13,400 (Tunis)AA AE 4,000 (London - max bus priority) 18,000 (London)AE AF 5,000 (L.A.) 26,000 (U.S.)AM AE 6,000 (London - busway) 30,000 (Putra Kuala Lumpur –theoretical estimate 7,500 (Adelaide)A only)AN 9,000 (SE Busways)AL 10,000 (Ottawa Transitway)AN 11,000 (Curitiba)AD 11,500 (Goiania, Brazil)AN 15,000 (Quito Trolleybus)AN 15,100 (Curitiba, Eixo Sul)AN, AA 21,100 (Belo Horizonte, Brazil)AN 25,600 (Porto Alegre, Farrapos)AN 26,000 (Porto Alegre)AA 28,000 (Porto Alegre, Assis)AN 29,800 (Recife Caxanga, Brazil)AN 33,000 (Bogota)AN 34,900 (Sao Paulo 9 de Julho)AN 67,000 (Bogotá - TransMilenio)A Standard Rigids: Standard Rigids: 30 m Length Tram: 30 m Length LRT: 25 m Length TransLohr: . 10 min frequency 450 . 10 min frequency 450 . 10 min frequency 1,074 . 10 min frequency 942 . 10 min frequency 1,200 . 5 min frequency 900 . 5 min frequency 900 . 5 min frequency 2,148 . 5 min frequency 1,884 . 5 min frequency 2,040 . 2 min frequency 2,250 1 min frequency . 2 min frequency 2,250 1 min frequency . 2 min frequency 5,370 1 min frequency . 2 min frequency 4,710 1 min frequency . 2 min frequency 5,100 1 min frequency 4,500 4,500 10,740 9,420 10,200 Tri-Axel Rigids: Tri-Axel Rigids: 40 m Length Tram: 40 m Length LRT: 32 m Length TransLohr: . 10 min frequency 540 . 10 min frequency 540 . 10 min frequency 1,290 . 10 min frequency 1,416 . 10 min frequency 1,380 . 5 min frequency 1080 . 5 min frequency 1080 . 5 min frequency 2,580 . 5 min frequency 2,832 . 5 min frequency 2,760 . 2 min frequency 2,700 1 min frequency . 2 min frequency 2,700 1 min frequency . 2 min frequency 6,450 1 min frequency . 2 min frequency 7,080 1 min frequency . 2 min frequency 6,900 1 min frequency 4,400 4,400 12,900 14,160 13,800 Articulated: Articulated: 54 m Length Tram: 54 m Length LRT: 39 m Length TransLohr: . 10 min frequency 660 . 10 min frequency 660 . 10 min frequency 2,112 . 10 min frequency 2,112 . 10 min frequency 1,740 . 5 min frequency 1,320 . 5 min frequency 1,320 . 5 min frequency 4,224 . 5 min frequency 4,224 . 5 min frequency 3,480 . 2 min frequency 3,300 . 2 min frequency 3,300 . 2 min frequency 10,560 . 2 min frequency 10,560 . 2 min frequency 8,700 . 1 min frequency 6,600 . 1 min frequency 6,600 . 1 min frequency 21,120 . 1 min frequency 21,120 . 1 min frequency 17,400

Double Articulated: 72 m Length Tram: 74 m Length LRT: 46 m Length TransLohr: . 10 min frequency 1,080 . 10 min frequency 3,180 . 10 min frequency 3,600 . 10 min frequency 2,070 . 5 min frequency 2,160 . 5 min frequency 6,360 . 5 min frequency 7,200 . 5 min frequency 4,140 . 2 min frequency 5,400 . 2 min frequency 15,900 . 2 min frequency 18,000 . 2 min frequency 10,350 . 1 min frequency 10,800 . 1 min frequency 31,800 . 1 min frequency 36,000 . 1 min frequency 20,700

Page A.2 10-0477-01-2106689A PARSONS BRINCKERHOFF

Appendix A Mode comparison summary

Bus on street Bus Rapid Transit (BRT) Trams on street Light Rail Transit (dedicated) TransLohr Capital expenditure (average per kilometre in AUD) Indicative range $0.1m-3m Indicative range $1m-32m Indicative range $10m-100m Indicative range $10m-100m Indicative range $10m-50m

<$1.75m (London)BT $0.39m-$0.78m (rapid bus)AF 5 $10m (Yarra Trams – ballpark figure, includes overhead $17.5m – $78m (London)BT $46.1m (Paris)BI AG power cabling, stations & services etc. does not include new BY BD $1.8m (Porto Alegre Busways) W $569.32m (San Francisco, USA) Third Street Extension $10.8m (Shanghai 2009) sub station)

$2.2m (LPT exclusive corridor)B 4 $192.32m (Seattle, USA) BY Central Link projects $7.45m (Claremont Farrand, France)BB $12m-$24m (trams – double track. Rising to around $72m if $2.4m - $4.8m (max priority - London)AE 7 substantial lengths of elevated track or tunnel are required. $148.57m (New Jersey, USA) BY W AE 7 Costs include depot, workshops, rolling stock & infrastructure) BY $2.4m - $48m (busway - London) $146.46m (Pittsburgh, USA) B 4 $16m (Tunis, covers planning & construction costs, technical BY (LPT shared, on median of arterials) AN 5 (Los Angeles, USA) Metro Gold Line east $3.4m equipment & rolling stock) $113.60m BV extension $3.4m (Eugene, USA) E $18m - $22m (Gold Coast estimate) BY B 4 $95.25m (Houston, USA) North Corridor $4.1m (LPT shared, outside arterial lanes) AF 5 $23m-$78m (indicative range) BY B 4 $81.85m (Houston, USA) Southeast Corridor $3.2m (LPT shared, on one side of road) AE 7 $23.4m (Manchester Metrolink) BY B 4 $70.33m (St Paul/Minneapolis, USA) $2.8m (LPT Greenfield exclusive corridor) AE 7 $24m - $106m (indicative range) BY AE 7 $59.34m (Denver, USA) Gold Line $3.8m (busway - London) $24-$32m (Stockholm, Sweden), BY AF 5 $53.72m (Phoenix, USA) $3.9m-$42.7m (indicative range, busway) AG $24.2m (Tunis) BY AM 5 $52.19m (Charlotte, USA) $7.4m (Bogota Phase I) $27m (US average – covers planning & construction costs, BY AG AN 5 $49.93m (Dallas, USA) $9.5m (Bogotá TransMilenio phase 1) technical equipment & rolling stock) BY BV Y 7 $46.62m (Sacramento, USA) $9.9m (Ottoway, Canada) $35.2m (Montpellier, France) BY A O AE 7 $43.74m (Denver, USA) West Corridor $11.2 (LPT Sydney) $40m (Tramlink extensions, London) BY BV $37.66m (Salk Lake City, USA) $17.8m (Los Angeles, USA) $63m (PUTRA – Kuala Lumpur: elevated, driverless, covers planning & construction costs, technical equipment & rolling $34.33m (Denver, USA) BY Southeast Corridor $16.6m (Bogota Phase II – difference primarily due to AN 5 stock) BY increased investment in public space & infrastructure $25.91m (Denver, USA) East Corridor AM 5 BT improvements) $26m - $35m (London) BY (Virginia, USA) BY $23.39m $22.4m (SE Busways, Brisbane – fully grade separated, $51.72m (Portland, USA) Portland Mall, South Corridor BY AI $93.57m (Orange, County, USA) tunnels/viaducts, stations) BY $150.33m (Portland, USA) Milwaukie BY CH $197.6m (Honolulu, USA) Elevated LRT $104 (Northern Busway, Brisbane) BZ $16.96m (Portland, USA) Downtown Line CA CH $50m (Dubai, UAE) , consists of fully automated, elevated $158 (Inner Northern Busway, Brisbane) BZ $19.64m (Portland, USA) Downtown Line Extension and underground track Y 7 $27.2m (av. of 22 automated guided systems in US) AE 5 $31.3m (Orange Line BRT –L.A) Ground based power supply BT 1999 $1.75-$35.0m (London) $33m (Bordeaux, France)BU Cost per Vehicle CK $187,500- $437,500 (CNG,LPG)AC 5 $187,500- $437,500 (CNG,LPG)AC 5 $2.9m (double-articulated) average 2005-2006AZ 5 $3.9m (Calgary, Canada) $3.1m (Claremont Farrand 2006)BB

$250,000- $500,000 (Hybrid Electric)AC 5 $250,000- $500,000 (Hybrid Electric)AC 5 $3m (Sydney estimates)A & AK $3.15m (Shanghai/Tianjin, China)CC- $312,000 (single decker)W $312,000 (single decker)W $3.2m (Madrid)BA 8

$384,000 (double decker)W $384,000 (double decker)W $3.5m (modern low-floor)AI

$480,000 (articulated single decker)W $480,000 (articulated single decker)W $3.4m (articulated) average 2005-2006AZ 5

$550,000 - $800,000 $550,000 - $800,000 $5.3m (1-level cab) average 2005-2006AZ 5 (18 mtrs, articulated, low-floor, standard, diesel or CNG)AF, AU 5 (18 mtrs, articulated, low-floor, standard, diesel or CNG)AF, AU 5

$780,000 - $1.2m $780,000 - $1.2m Ground based power supply (18 mtrs, articulated, low-floor, stylised (looks like LRT), diesel (18 mtrs, articulated, low-floor, stylised (looks like LRT), diesel or 1999 BU AF, AU 5 AF, AU 5 (Bordeaux, France) or CNG) CNG) $3.55

$1m (high capacity buses, Sydney estimates)AK $960,000 (optically guided, articulated single decker)W

$1.25m-$1.88m (Fuel Cell)AC 5 $1m (high capacity buses, Sydney estimates)AK

$1.2m - $2m (specialised BRT vehicles – e.g. Civis by Irisbus in Las Vegas)AU 5 $1.2m - $2m (18 mtrs BRT vehicle with guidance, internal combustion, - electric or hybrid) AV

$1.25m-$1.88m (Fuel Cell)AC 5 $2.16m (French GLT articulated single decker)W

PARSONS BRINCKERHOFF 10-0477-01-2106689A Page A.3

Appendix A Mode comparison summary

Bus on street Bus Rapid Transit (BRT) Trams on street Light Rail Transit (dedicated) TransLohr Operational expenditure (cost per vehicle kilometre, cost per vehicle hour, cost per passenger kilometre) Typical Australian cost $3-$4W Typical Australian cost $5-$15 Typical Australian cost $5-$15

$7.00 - $14.00/vkm (London)BT $4.50/vkm (not ‘next generation’ BRT)AH $12/vkm (Tramlink - London)AE 7 $12/vkm (Tramlink - London)AE 7 W W $9-$19/vkm (bus)AE 7 $13/vkm $13/vkm AH AH $7.00 - $14.00/vkm (London)BT $14/vkm $14/vkm $8.35/vkm (London)BT $21.00/vkm (DLR, London)BT $162 per hr (Dallas, USA)BP $335 per hr (Dallas, USA) BP $130 per hr (Denver, USA) BP $205 per hr (Denver, USA) BP $43 per hr (Los Angeles, USA) BP $724 per hr (Los Angeles, USA) BP $238 per hr (Pittsburgh, USA) BP $378 per hr (Pittsburgh, USA) BP $173 per hr (San Diego, USA) BP $151 per hr (San Diego, USA) BP $184 per hr (San Jose, USA) BP $335 per hr (San Jose, USA) BP $53 per hr (Calgary, Canada) BP $122 per hr (Calgary, Canada) BP Cost per Passenger Km Cost per Passenger Km Cost per Passenger Km $0.73 (Santiago, USA)BW $0.68 (Portland, USA) BW $0.33 (Santiago, USA)BW $1.27 (St Louis, USA) BW $0.91 (Dallas, USA) BW $0.39 (St Louis, USA) BW $0.93 (Los Angeles, USA )BW $1.66 (Buffalo, USA) BW $0.58 (Los Angeles, USA )BW $0.98 (Portland, USA) BW $0.31 (Stockholm, Sweden)BX $0.73 (Sacramento, USA) BW $1.02 (Sacramento, USA) BW $0.93 (Baltimore, USA) BW $1.43 (Dallas, USA) BW $1.53 (San Jose, USA) BW $1.02 (Baltimore, USA) BW $1.18 (Denver, USA) BW $1.02 (Denver, USA) BW $1.39 (San Jose, USA) BW $1.51 (Buffalo, USA) BW $0.26 (Stockholm, Sweden)BX Operating speed (includes loading at stations) Maximum Operating Speed: Maximum Operating Speed: Maximum Operating Speed: Maximum Operating Speed: Maximum Operating Speed: 80 km/h – 100 km/h 100km/h 70 km/h 100 km/h – 110 km/h 70 km/h 5 50 km/h on GLPS (Bordeaux, France) 10-14 km/hr (bus)AE 15-22 km/hr (busway)AE 20-22 km/hr AE (Tram) 18-40 km/hr AE (LRT) 14-18 km/hr (max priority)AE 14-18 km/h (maximum priority)BT 8-10 km/hr (Adelaide)BF 25 km/hr (Adelaide)BH 15-22 km/h (Express) 15-22 km/h (busway)BT 15-22 km/h (London)BT 30-50 km/h A 22-29 km/h (full BRT)AA 18-40 km/h (London)BT 30-60 km/hA 45-50 km/hr (SE Busway) AL 55-60 km/hr (O-Bahn) BE Turning radii 10 m Radius 10 m – 15 m Radius 18 m – 25 m Radius 25 m Radius 10.5 m Radius Power source Diesel Diesel Generally: 550-800V DC Generally: 1.5KV Generally: 600-750V DC Compressed Natural Gas Compressed Natural Gas Fuel Cell Fuel Cell Overhead Power Supply Overhead Power Supply Overhead Power Supply Hydrogen Hydrogen Ground Level Power Supply (GLPS) Ground Power Supply WiPost (non Catenary) Trolley Bus (Overhead wires) Overhead Power Supply Battery Trolley Bus (Overhead wires) Diesel

Timetable and technology reliability Low - Medium Medium – Maximum Priority Medium – with traffic Good – Segregated corrid Medium – with traffic Good – Grade Separation Good – Segregated corridor Good – Segregated corridor

Page A.4 10-0477-01-2106689A PARSONS BRINCKERHOFF

Appendix A Mode comparison summary

Bus on street Bus Rapid Transit (BRT) Trams on street Light Rail Transit (dedicated) TransLohr Technology maturity Very High maturity with conventional buses for Very high for conventional buses on a BRT Very high maturity with low floor, ultra low floor Very high maturity with low floor, ultra low floor Low to moderate maturity for TransLohr vehicles. standard and articulated vehicles. system. and high floor vehicles. and high floor vehicles. There have been several issues dealing with Very high maturity for double articulated high floor For mechanically guided busways the technology derailments in several systems including the vehicles, however, low floor double articulated is very high however, low floor articulated recently opened systems in Shanghai (2009). vehicles are not as common. mechanically guided busway vehicles has a low There is also a high level of maturity for hybrid maturity and current problems exist with buses in the United States and Europe, Hybrid operational speeds and vibrations. buses in Australia are still emerging. Optically guided busway vehicles have a moderate maturity with some initial problems. Integration with the pedestrian and public realms Ability for pedestrians to cross at grade: Ability for pedestrians to cross at grade: Ability for pedestrians to cross at grade: Ability for pedestrians to cross at grade: Ability for pedestrians to cross at grade: . Yes. . Yes if slow speed operation . Yes. . Yes . Yes . Preferably No for higher speed operation . Preferably No for higher speed operation . Preferably No when in dedicated higher . No if BRT is guided . Preferably No if LRT level of service is greater speed corridors. . No if BRT level of service is greater than than 30 trams per hour. 30-45 buses per hour. Bus stops located on street, indented bus bays or Bus stations located central medians require Tram stations located in central medians require LRT stations located in central medians require TransLohr stations located in central medians within interchanges. Easily integrated with signalised crossings, platform heights are similar signalised crossings; platform heights are slightly signalised crossings; platform heights are slightly require signalised crossings; platform heights are pedestrian realm as boarding height is similar to to regular kerb heights for low floor vehicles, high higher than regular kerb heights (300-350 mm) for higher than regular kerb heights (300-350 mm) for similar height to regular kerb heights (250 mm) regular kerb height. floor vehicles either require steps within the low floor trams. Trams systems with side running low floor trams. LRT systems with high floor trams for TransLohr vehicles. TransLohr systems with vehicle and wheelchair lifts or high floor platforms lanes can be incorporated into pedestrian require high floor platforms with ramps or side running lanes can be incorporated into with ramps or wheelchair lifts. BRT systems with footpaths with raised platform areas at stations. wheelchair lifts. pedestrian footpaths with raised or kerb height side running lanes can be directly integrated into Tram stations can also be located in traffic lanes LRT with dedicated running lanes in streets platform areas at stations. pedestrian footpath. with a shared pedestrian and traffic boarding generally require fencing between the track to The single central rail used to guide TransLohr area. Platform heights remain the at 300-350 mm prevent pedestrians from crossing in non vehicles has been known to cause issues with with the road travel lanes raising to this height at dedicated areas. This can create barriers on cyclists and wheelchairs crossing the track due tram stations. streetscapes and can be visually un appealing. to the groves required for the vehicle’s wheels. Non accessible tram stations can also be located Other forms of dedicated on in street running in traffic lanes, signals stop traffic in both travel include raised kerbs or dedicated road markings directions when a tram stop to let passengers out. to prevent regular street traffic from entering the corridor. Visual Amenity Increased localised pollution from diesel vehicles Guided BRT can require guideways which can be Over head wires if non-ground based power Over head wires if non-ground based power Overhead wires. including particulate build-up on surrounding visually unattractive. supply or internal combustion engine technology supply or internal combustion engine technology Corridor surface can either be bituminised track, structures Oil stains, black marks and scuffing can occur chosen. chosen. concrete, paved or grassed depending on the Greater number of vehicles operating if high around bus stops and stations. Corridor surface can either be bituminised track, Corridor surface can either be bituminised track, application. capacity is required High floor bus stops can cause visual barriers concrete, paved or grassed. concrete, paved, open track or grassed Options do exist for WiPost operation which is Buses can cause damage to road surfaces to across the corridor. Ground level power supply system’s occur in depending on the application. overhead power supply without wires. Power is create uneven pavements. several cities, Bordeaux, France has the largest drawn from light posts/poles and are evenly Corridor surface for mechanically guided BRT’s BU Oil stains, black marks and scuffing can occur can be either open track or grassed. Optically system and has a reliability of 98.4% reliability , spaced along the line. Each unit of the vehicle around bus stops. guided BRT systems generally require a concrete minimal visual amenity and corridor surface can has a long horizontal pole located on the roof of or bitumen surface. be the same as regular tram systems. the tram that is in direct contact with the WiPost, ensuring the tram has continuous power supply without the need for wires.

PARSONS BRINCKERHOFF 10-0477-01-2106689A Page A.5

Appendix A Mode comparison summary

Bus on street Bus Rapid Transit (BRT) Trams on street Light Rail Transit (dedicated) TransLohr Value uplift and redevelopment catalyst There has been little research in relation to the 20% increase in property values (Brisbane) BO 25% increase in property values (San Diego, 2.6% increase in property values over 2009/2010 Since TransLohr technology is relatively new, BK BL increase of standard bus services on property 5-10% increase in residential property values USA) . around rail stations in Brisbane . research on property value uplift for this values. This is because the fluctuation between within 300 m of BRT Stop and 3-26% increase in 10% increases in property values (Portland, USA 6.7% increases in property values (Boston, USA technology is minimal. It can be assumed that the level of service provided to residential areas retail values within 150 m of BRT stop (Seoul, 1992)BM. 1994)BM this form of technology however, will attract BQ can change from suburb to suburb or city to city. Korea) . 20% higher property values (Newcastle upon 10-15% higher rental returns (San Francisco, similar levels of property uplift as trams and LRT However, it should be noted that Australia real 15-20% within walking distance of BRT stations, Tyne, UK 2004)BM. USA 1996) BM. depending upon the application and perception estate agents often advertise that a property is of its users. however -3% to -4.4% within 150 m of BRT trunk 10% higher property values (Manchester, UK 10.5% increases in property values (Washington, located close to public transport services. line (Bogota, Columbia)BR. BM BM Therefore, it could be assumed that there is some 2004) . USA 1999) . value uplift for properties that are located with Pittsburgh East Busway prompted $196 m of 7% higher residential rental returns (Strasburg, 19-33% increase in property values (Holland, additional development along the corridor BM BM proximity of good, reliable and frequent bus BS France 2004) . 2006) . services. (Pittsburgh, USA) . 10%-15% higher office rental returns (Strasburg, 25% increase in property values (Dallas, USA France 2004)BM. 2000)BM. 15%-20% higher office rental returns (Freiburg, Germany 2004)BM. 50% higher office rental returns (Bremen, Germany 2004)BM. 40% increase of property values 1 block from tram stations (Portland, USA Downtown)CD. Portland Metropolitan Express (15 miles/ 32 stations, plus plans for 18 miles expansion): Since 1986, $1.9 billion in property. development in the immediate vicinity of the line.BM. St Louis, Missouri (opened 1993, 18 miles/ 18 stations): to date, development spurred by transit system totals $530 million and includes major projects. A $1.5 billion expansion of LRT is expected to have a $2.3 billion impact on business salesBM.

1. Based on standard Australian built buses in service around Australia 2. Based on standard Australian 3A. Based on Bombardier’s Flexity 2 specifications 3B. Based on Siemens Light Rail Specifications 4. Based on TransLohr’s Specification 5. GLPS – Ground Level Power Supply 6. Based on Portland 20 m LRT vehicles 7. Based on Charlotte 27 m LRT vehicles

Sources: A: F6 Public Transport Use Assessment B: PB LPT Tway design paper C: CityRail website D: Victoria Transport Policy Institute August 2006, Rail Transit in America Comprehensive Evaluation of Benefits, P34 E: Gold Coast Light Rail Feasibility Study 2005 F: Sydney Buses G. Glazebrook & Associates February 2005, Report to City of Sydney, Integrated Transport Strategy – Mass Transit for CBD and Inner Sydney. H. The Sydney Light Rail Company. Light Rail in Sydney-Issues and Perspectives. April 1997 I: Transportation Research Board. Bus Rapid Transit: Why more communities are choosing Bus Rapid Transit. 2001 J. Sinclair Knight Merz. Liverpool to Parramatta Rapid Bus Transitway Environmental Impact Statement Volume 1. August 2000 K. Transport Infrastructure Development Corporation. South West Rail Link Concept Plan and Environmental Assessment. Submissions Report. May 2007 L. Rouen, France. Brief. Teor Optically Guided Bus M. MTR Corporation 2006 Annual Report. N. NSW Auditor General Report Performance Audit, Liverpool to Parramatta Bus Transitway, 2005 O. Parsons Brinckerhoff. Central Sydney Light Rail Transport Operations Study. January 2004 P. KCRC Corporation 2006 Annual Report Q. Hass-Klau Carmen et.al Bus or Light Rail: Making the right choice. December 2003

Page A.6 10-0477-01-2106689A PARSONS BRINCKERHOFF

Appendix A Mode comparison summary

R. Alternative Propulsion Concepts. 2005 UITP Conference Presentation S. Tramways & Urban Transit December 2002. Bordeaux: Fronting the French tramway revolution T. Transport for London (Docklands Light Railway) Website U. Bus Rapid Transit Superior to Light Rail: US GAO Report Results W. PB Lockerbie Light Rail Study X. RailCorp Civil Engineering Standards 2007 Y. Transek Consultants, Comparison of costs between Bus, PRT, LRT and metro/rail. Z. Piers Brogan Presentation at ITLS (Brisbane Airport case study) AA. PB Reference Library – Introduction to BRT AB. PB Reference Library – Operational Plan AC. PB Reference Library – Technology AD. Hensher BRT or Light Rail AE. PB Transport for London presentation. AF. PB BRT Cost Comparison presentation (Cliff Henke) AG. World Bank: Cities on the move: a world bank urban transport strategy review (figures are given as US$ in Sept 2000. These have been converted to Australian at the rate of AUS$1 = US$0.55) AH. http://ite-espanol.org/meetcon/2005AM/Evans_Tues.pdf AI. Dick Fleming Transitway presentation AK. PB Central Sydney Light Rail Transport Operations Study AL. Ken Gosselin, Busway Experience Downunder presentation (McCormick Rankin Corporation) AM. National Bus Rapid Transit Institute, Applicability of Bogota’s TransMilenio BRT System to the United States AN. Mass Transit Options - Sustainable Transport: A Sourcebook for Policy-makers in Developing Cities Module 3a AO. NSW Gov. – Metro Lines: A Part of Sydney’s Future? 2006 AP. 2006 RailCorp Rail Development Sectorisation. A Compendium of City Rail Travel Statistics Fifth Edition, April 2006 AQ. Breakthrough Technology Institute. Bus Rapid Transit. A cost effective sustainable mobility solution AR. Based on $2.1 billion CityRail annual operating figure cost divided by the total train kilometres travelled (34,741,200 kms travelled in 2005 Compendium of City Rail Travel Statistics Fifth Edition, April 2006) AS. Sydney Light Rail Technical Details AT. UITP 5th Bus Conference, 2007, “Results from the UITP Working Group “High Capacity Surface Systems” AU. FTA (2004), Characteristics of Bus Rapid Transit for Decision-Making. AV. TCRP Report 90 Bus Rapid Transit v2 Implementation Guidelines 2003 AW. Vehicle Catalog: A Compendium of Vehicles and Powertrain Systems for Bus Rapid Transit Service 2006 Update AX. http://www.2getthere.eu/Bus_Transit/Specifications/Technical_Specification AY. Reconnecting America – Transit Technologies Worksheet AZ. http://www.apta.com/research/stats/rail/railcost.cfm BA. http://www.railway-technology.com/projects/madrid-light-rail/ BB. http://www.emta.com/article.php3?id_article=314 BC. http://railforthevalley.wordpress.com/2009/10/26/tram-on-tires-guided-light-transit-glt-the-ultimate-guided-bus/ BD. http://www.railwaysafrica.com/2009/02/translohr-for-shanghai/ BE. Based on current public timetable operation BF. Based on inner city public timetable operation BH. Based on dedicated ROW corridor public timetable operation BI. http://railforthevalley.wordpress.com/2009/10/ BJ. http://www.shanghai.gov.cn/shanghai/node17256/node18151/userobject22ai31185.html BK. http://www.transportroundtable.com.au/smart/hensher.pdf BL. http://www.brisbanetimes.com.au/queensland/brisbane-home-buyers-on-the-right-track-20100506-udco.html BM. http://www.infrastructureaustralia.gov.au/public_submissions/published/files/486_propertycouncilofaustralia_SUB2.pdf BN. http://www.smh.com.au/nsw/light-rail-to-push-up-house-prices-20100312-q469.html BO. http://www.nctr.usf.edu/jpt/pdf/JPT%209-3S%20Currie.pdf BP. http://www.calgarytransit.com/pdf/brt_report.pdf BQ. Cervero, R, Kang, C D (2009), Bus Rapid Transit Impacts on Land Uses and Land Values in Seoul, Korea, UC Berkeley Center for Future Urban Transport, California, USA BR. http://www.lincolninst.edu/pubs/dl/1353_671_Rodriguez%20Mojica%20Final.pdf BS. http://www.nbrti.org/docs/pdf/BRT%20and%20land%20use_97ver_508.pdf BT. Transport for London BU. Bordeaux – APS Ground Power Supply System, Parsons Brinckerhoff, October 2005 BV. Cameron Road Corridor Study, Tauranga City Council, Beca Infrastructure LTD, March 2010 BW. http://www.lightrailnow.org/facts/fa_lrt02.htm BX. http://www.jpods.com/JPods/004Studies/CostPerMileOperations_UWa.pdf BY. http://www.prtstrategies.com/files/LRT_Costs.pdf CA. http://www.lightrailnow.org/news/n_newslog002.htm#DBi_20050114 CB. http://www.gobrt.org/CaseStudies.pdf

PARSONS BRINCKERHOFF 10-0477-01-2106689A Page A.7

Appendix A Mode comparison summary

CC. http://www.chinaeconomicreview.com/today-in-china/2010_03_26/On_the_right_track.html CD. http://www.city.urbana.il.us/urbana/community_development/planning/archives/MTD_Tram_Study.pdf CE. http://www.emta.com/IMG/pdf/emta_news_23.pdf CF. http://www.thetransportpolitic.com/page/11/ CG. Railway Gazette International (2009), Primove Catenary-free induction tram, http://www.railwaygazette.com/news/single-view/view/10/primove-catenary-free-induction-tram.html CH. http://www.lightrailnow.org/news/n_newslog2009q2.htm#BRB_20090605 CI. http://www.lohr.fr/transport-public_gb.htm CJ. http://www.llbc.leg.bc.ca/public/pubdocs/bcdocs/368601/attach2.pdf CK. http://www.calgarytransit.com/html/technical_information.html CL. http://glassborocamdenline.com/images/uploads/AppendixD.pdf

Page A.8 10-0477-01-2106689A PARSONS BRINCKERHOFF

Appendix B

Operating scenario and costs

Appendix B Operating scenario and costs

Operating scenario and costs

Assumptions Distance from Stirling Station to Stirling City Centre 1.00 km Number of Weekdays 251 per annum

Distance from Stirling City Centre to Glendalough Station 2.40 km Number of Saturdays 52 per annum

Number of Sundays and Public Holidays 62 per annum

Average Running Speed BRT 17.50 km/h Tram/TransLohr 20.00 km/h Peak hour percentage of daily trips 10% LRT 25.00 km/h Total daily trips as a percentage of weekdays 100% Weekdays Optimum Capacity Seated Passengers 100% of seated capacity 75% Saturdays Standing Passengers 75% of standing capacity 50% Sundays

Seated Standing Total Cost per Vehicle Operation Cost per km Corridor Capital Cost per km Standard Bus 45 30 75 $400,000 $5.00 $7.50m Tri-Axel Bus 55 35 90 $450,000 $5.50 $7.50m Articulated Bus 65 55 120 $600,000 $6.00 $7.50m

30 m Tram 64 115 179 $4,500,000 $10.00 $10.00m 40 m Tram 72 143 215 $5,500,000 $11.00 $12.00m 54 m Tram 58 296 354 $6,000,000 $12.00 $14.00m 72 m Tram 90 440 530 $6,500,000 $13.00 $15.00m

20 m LRT 30 127 157 $3,900,000 $10.00 $15.00m 30 m LRT 68 168 236 $5,500,000 $11.00 $17.50m 40 m LRT 100 210 310 $5,750,000 $12.00 $20.00m 74 m LRT 180 420 600 $6,000,000 $13.00 $22.50m

25 m TransLohr 60 110 170 $2,900,000 $7.25 $8.00m 32 m TransLohr 80 150 230 $3,190,000 $7.25 $9.00m 39 m TransLohr 100 190 290 $3,480,000 $7.25 $10.00m 46 m TransLohr 120 225 345 $3,770,000 $7.25 $11.00m

Please note: Maintenance cost have not been calculated for the infrastructure or the vehicles. Spare vehicles have not been included in the calculation for the number of vehicles. A 10% spare ratio should be acceptable

PARSONS BRINCKERHOFF 10-0477-01-2106689A Page B.1

Appendix B Operating scenario and costs

Bus/Tram/LRT/TransLohr Headways and Required Vehicles Vehicles per Hour Headways (minutes) Required number of vehicles

Optimum

Passenger

Total Load 2000pp/h 2500pp/h 3000pp/h 3500pp/h 4000pp/h 5000pp/h 6500pp/h 2000pp/h 2500pp/h 3000pp/h 3500pp/h 4000pp/h 5000pp/h 6500pp/h 2000pp/h 2500pp/h 3000pp/h 3500pp/h 4000pp/h 5000pp/h 6500pp/h Standard Bus 68 30 37 44 52 59 74 96 2 2 1 1 1 1 1 16 21 25 29 33 41 53 Tri-Axel Bus 81 25 31 37 43 49 62 80 2 2 2 1 1 1 1 14 17 21 24 27 34 44 Articulated Bus 106 19 24 28 33 38 47 61 3 3 2 2 2 1 1 10 13 16 18 21 26 34

30 m Tram 150 13 17 20 23 27 33 43 5 4 3 3 2 2 1 7 8 10 12 13 17 22 40 m Tram 179 11 14 17 20 22 28 36 5 4 4 3 3 2 2 6 7 8 10 11 14 18 54 m Tram 280 7 9 11 13 14 18 23 8 7 6 5 4 3 3 4 5 5 6 7 9 12 72 m Tram 420 5 6 7 8 10 12 15 13 10 8 7 6 5 4 2 3 4 4 5 6 8

20 m LRT 125 16 20 24 28 32 40 52 4 3 3 2 2 2 1 7 9 11 12 14 18 23 30 m LRT 194 10 13 15 18 21 26 34 6 5 4 3 3 2 2 5 6 7 8 9 11 15 40 m LRT 258 8 10 12 14 16 19 25 8 6 5 4 4 3 2 3 4 5 6 7 9 11 74 m LRT 495 4 5 6 7 8 10 13 15 12 10 8 7 6 5 2 2 3 3 4 4 6

25 m TransLohr 143 14 18 21 25 28 35 46 4 3 3 2 2 2 1 7 9 11 12 14 18 23 32 m TransLohr 193 10 13 16 18 21 26 34 6 5 4 3 3 2 2 5 7 8 9 11 13 17 39 m TransLohr 243 8 10 12 14 16 21 27 7 6 5 4 4 3 2 4 5 6 7 8 10 14 46 m TransLohr 289 7 9 10 12 14 17 23 9 7 6 5 4 3 3 4 4 5 6 7 9 11

Optimum frequency = 5 minutes

Page B.2 10-0477-01-2106689A PARSONS BRINCKERHOFF

Appendix B Operating scenario and costs

Bus/Tram/LRT/TransLohr Capital and Operational Costs

Cost for vehicles (excluding spare vehicles) Annual Operation Cost (excludes maintenance) Lifetime (25 years ) Capital and Operational Cost (excludes maintenance)

Capital Cost 2000pp/h 2500pp/h 3000pp/h 3500pp/h 4000pp/h 5000pp/h 6500pp/h 2000pp/h 2500pp/h 3000pp/h 3500pp/h 4000pp/h 5000pp/h 6500pp/h 2000pp/h 2500pp/h 3000pp/h 3500pp/h 4000pp/h 5000pp/h 6500pp/h

Standard Bus $25.5m $6.58m $8.23m $9.87m $11.52m $13.16m $16.45m $21.39m $3.23m $4.04m $4.85m $5.66m $6.47m $8.08m $10.51m $1,356m $2,104m $3,018m $4,099m $5,346m $8,338m $14,074m

Tri-Axel Bus $25.5m $6.15m $7.69m $9.23m $10.76m $12.30m $15.38m $19.99m $2.96m $3.69m $4.43m $5.17m $5.91m $7.39m $9.60m $1,035m $1,603m $2,297m $3,118m $4,064m $6,336m $10,691m

Articulated Bus $25.5m $6.27m $7.84m $9.41m $10.97m $12.54m $15.68m $20.38m $2.47m $3.08m $3.70m $4.31m $4.93m $6.16m $8.01m $670m $1,032m $1,475m $1,998m $2,602m $4,051m $6,829m

30 m Tram $34.0m $30.35m $37.94m $45.52m $53.11m $60.70m $75.87m $98.64m $2.91m $3.63m $4.36m $5.08m $5.81m $7.26m $9.44m $524m $799m $1,136m $1,534m $1,994m $3,096m $5,209m

40 m Tram $40.8m $31.09m $38.87m $46.64m $54.41m $62.19m $77.73m $101.05m $2.68m $3.35m $4.02m $4.69m $5.36m $6.70m $8.71m $419m $632m $893m $1,200m $1,555m $2,407m $4,040m

54 m Tram $47.6m $21.71m $27.14m $32.57m $38.00m $43.43m $54.29m $70.57m $1.87m $2.34m $2.81m $3.27m $3.74m $4.68m $6.08m $217m $312m $428m $566m $725m $1,106m $1,836m

72 m Tram $51.0m $15.68m $19.60m $23.52m $27.44m $31.37m $39.21m $50.97m $1.35m $1.69m $2.03m $2.36m $2.70m $3.38m $4.39m $133m $178m $234m $301m $377m $560m $912m

20 m LRT $51.0m $27.32m $34.15m $40.98m $47.81m $54.64m $68.30m $88.78m $3.49m $4.36m $5.23m $6.10m $6.97m $8.71m $11.33m $661m $1,005m $1,424m $1,920m $2,492m $3,866m $6,498m

30 m LRT $59.5m $24.87m $31.09m $37.31m $43.53m $49.75m $62.18m $80.84m $2.48m $3.09m $3.71m $4.33m $4.95m $6.19m $8.04m $339m $497m $689m $917m $1,179m $1,809m $3,016m

40 m LRT $68.0m $19.59m $24.49m $29.39m $34.28m $39.18m $48.98m $63.67m $2.03m $2.54m $3.05m $3.56m $4.07m $5.09m $6.61m $241m $339m $458m $599m $761m $1,151m $1,898m

74 m LRT $76.5m $10.63m $13.29m $15.95m $18.61m $21.27m $26.59m $34.56m $1.15m $1.43m $1.72m $2.01m $2.29m $2.87m $3.73m $127m $156m $191m $232m $280m $394m $613m

25 m TransLohr $27.2m $20.62m $25.78m $30.93m $36.09m $41.24m $51.56m $67.02m $2.22m $2.78m $3.33m $3.89m $4.44m $5.55m $7.22m $422m $644m $916m $1,236m $1,607m $2,495m $4,198m

32 m TransLohr $30.6m $16.79m $20.99m $25.19m $29.39m $33.58m $41.98m $54.58m $1.64m $2.06m $2.47m $2.88m $3.29m $4.11m $5.34m $247m $369m $517m $693m $896m $1,383m $2,316m

39 m TransLohr $34.0m $14.54m $18.18m $21.81m $25.45m $29.08m $36.35m $47.26m $1.31m $1.63m $1.96m $2.28m $2.61m $3.26m $4.24m $170m $247m $341m $452m $579m $886m $1,474m

46 m TransLohr $37.4m $13.23m $16.54m $19.85m $23.15m $26.46m $33.08m $43.00m $1.10m $1.37m $1.64m $1.92m $2.19m $2.74m $3.56m $134m $188m $254m $332m $422m $638m $1,053m

Lowest Cost Highest Cost Lowest Cost Highest Cost Lowest Cost Highest Cost

PARSONS BRINCKERHOFF 10-0477-01-2106689A Page B.3

Appendix C

Comparative operating characteristics

Appendix C Comparative Operating Costs

Average Time Distance Service Route From To Speed (Minutes) (km) (k/Ph) Adelaide O-Bahn O-Bahn Track Tea Tree Plaza End of O-Bahn 12 12 60.00 Track Non O-Bahn End of O-Bahn City 7 3.2 27.43 Track Track Whole trip Tea Tree Plaza City 20 15.2 45.60 Nth-West T-Way T65 Rouse Hill Parramatta 41 17 24.88 Town Centre T75 Rouse Hill Blacktown 25 11.3 27.12 Town Centre Liverpool to T80 Liverpool Parramatta 56 31 33.21 Parramatta T-Way Brisbane South East 111 City Eight Mile 27 16.5 36.67 Busway Plains Stn CityRail InterCity Northern Line Hornsby Central 36 34 56.67 CityRail Suburban ECRL Epping Chatswood 18 13 43.33 Adelaide Tram line Glenelg Tram City Glenelg 45 12.3 16.40 Adelaide Metro Bus 265 City Glenelg 35 12.6 21.60 Adelaide Heavy Rail Outer Harbor City Outer Harbor 39 21.9 33.69 Line Stn Sydney Light Rail CBD to Central Lilyfield 25 7.2 17.28 Lilyfield Perth Heavy Rail Joondalup Perth Clarkson Stn 32 33.2 62.25 Line Underground Melbourne Tram Segregated Burwood Vermont South 17 8 28.24 Route 75 track (centre of road, cross streets) On-street in- City - Spencer Burwood 57 15 15.79 traffic running St Whole trip City - Spencer Vermont South 74 23 18.65 St Melbourne Tram On-street in- North Prahran 32 7 13.13 Route 78 traffic running Richmond (Whole Trip) Melbourne Tram Segregated South St Kilda 10 4.5 27.00 Route 96 track (old Melbourne heavy rail alignment) On-street in- East Brunswick South 42 9.5 13.57 traffic running Melbourne via CBD St Kilda St Kilda Beach Whole trip East Brunswick St Kilda Beach 52 14 16.15 via City Source: Sydney Metro

PARSONS BRINCKERHOFF 10-0477-01-2106689A Page C.1

Appendix D

Stirling City Centre- Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Stirling City Centre

- Light Rail Feasibility Study

- Phase 2 TOD Trips Model Working Paper

October 2010

City of Stirling

Parsons Brinckerhoff Australia Pty Limited ABN 80 078 004 798 Level 27, Ernst & Young Centre 680 George Street SYDNEY NSW 2000 GPO Box 5394 SYDNEY NSW 2001 Australia Telephone +61 2 9272 5100 Facsimile +61 2 9272 5101 Email [email protected]

Certified to ISO 9001, ISO 14001, AS/NZS 4801 2106689A-PR_2812 A+ GRI Rating: Sustainability Report 2009

Revision Details Date Amended By Original

Signed: ......

Reviewer: Dick Fleming ......

Signed: ......

Approved by: Dick Fleming ......

Signed: ......

Date: 21 October 2010 ......

Distribution: ......

2106689A-PR_2812 Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Contents

Page number

1. Introduction 1

1.1 Background 1 1.2 TODTrips model 1 1.3 Report outline 2

2. Model scope 3

2.1 Introduction 3 2.2 Study area and zone system 3 2.3 Representation of network, services and access among zones 6 2.4 Public transport scenarios for 2031 7 2.4.1 Scenario 1 – Base case (Bus) 8 2.4.2 Scenario 2 – Street Car 9 2.4.3 Scenario 3A – LRT 10 2.4.4 Scenario 3B – BRT 11 2.4.5 Scenario 4 – LRT (single sided) 12

3. Land use considerations 15

3.1 Overview 15 3.2 Study area land use 15 3.2.1 Methodology 15 3.2.2 2031 land use development 15 3.3 External zones 19

4. Trip generation and distribution 21

4.1 Overview 21 4.2 Trip generation 21 4.3 Trip distribution 22 4.3.1 Overview 22 4.3.2 Internal trips 23 4.3.3 External trips 23

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page i

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Contents (Continued)

Page number

5. Mode choice model 25

5.1 Model structure 25 5.1.1 Mode choice model for internal trips (I-I) 25 5.1.2 Mode choice model for external trips (I-E and E-I movements) 26 5.2 Travel attributes 27 5.2.1 Overview 27 5.2.2 Car travel 28 5.2.3 Public transport travel 29 5.2.4 Walking and cycling 29 5.3 Generalised costs of travel 30 5.3.1 Formulating generalised costs to represent different trip segments between origin and destination 30 5.3.2 Weighting trip segments used in calculating generalised costs of travel 32

6. Assessment result of five transit modal scenarios 35

6.1 Internal trips – mode share and ridership estimates 35 6.2 External trips – mode share and ridership estimates 36 6.3 Combined internal and external trips - mode share and ridership estimates 38

7. Conclusions 41

Page ii 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Contents (Continued)

Page number

List of tables Page number

Table 2.1 Existing Public Transport’s operating environment for Stirling 7 Table 2.2 Scenario base (Bus) operating environment 8 Table 2.3 Scenario S2 (Street Car) operating environment 9 Table 2.4 Scenario 3A (LRT) operating environment 10 Table 2.5 Scenario 3B (BRT) operating environment 11 Table 2.6 Scenario 4 (LRT sided) operating environment 12 Table 3.1 Distribution of GFAs across different land uses for Stirling study area in 2031 16 Table 3.2 Average household size for different types of dwellings 17 Table 3.3 Floor space in m2 per employee from Syme Marmion & Co report and RTA NSW Guideline 17 Table 3.4 Adjusted average number of employees per 100 m2 GFA based on SMC and RTA NSW’s guideline 18 Table 3.5 Population and employment estimates for different land use categories in Stirling study area in 2031 18 Table 3.6 Population and employment estimates for Stirling Study Area in 2031 18 Table 3.7 List of 33 external zones of TODTrips model for Stirling Study 19 Table 4.1 Adjusted Average Daily Trip per 100 m2 GFA based on SMC and RTA NSW’s Guideline21 Table 4.2 Trip generation estimates for different land use categories in Stirling Study area in 203122 Table 4.3 Average Daily Trip Generation estimates for Stirling Study area in 2031 22 Table 4.4 Distribution of Average Daily Trips generated from Study Area in 2031 24 Table 5.1 Attributes of main travel mode used in the Stirling Study 28 Table 5.2 Generalised cost weights 33 Table 6.1 Low car use – Mode share for internal trips 35 Table 6.2 High car use scenario – Mode share of internal trips 35 Table 6.3 Low car use scenario – ridership share among different transport modes for internal trips36 Table 6.4 High car use scenario – ridership share among different transport modes for internal trips36 Table 6.5 Distribution of regional access and egress by external trips into and out of study area (I-E and E-I movements) 37 Table 6.6 Modal split of external trips using local transit services 37 Table 6.7 Ridership estimates of external trips (I-E and E-I movements) using local transit services38 Table 6.8 Low car use – Mode share for combined internal and external trips 38 Table 6.9 High car use – Mode share for combined internal and external trips 39 Table 6.10 Low car use – ridership estimates for combined internal and external trips 39 Table 6.11 High car use – ridership estimates for combined internal and external trips 39

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page iii

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Contents (Continued)

Page number

List of figures Page number

Figure 2.1 Stirling study area 4 Figure 2.2 TODTrips zone system 5 Figure 2.3 Stop locations of existing public transport service 8 Figure 2.4 Stop locations of new local bus service in Scenario 1 9 Figure 2.5 Stop locations of street car service in Scenario 2 10 Figure 2.6 Stop locations of LRT service in Scenario 3A 11 Figure 2.7 Stop locations of BRT service in Scenario 3B 12 Figure 2.8 Stop locations of LRT (sided) service in Scenario 4 13 Figure 3.1 Study area by groups of precincts 16 Figure 4.1 Trip distribution modeling – scope and assumption used for Stirling Study 23 Figure 5.1 Mode choice model structure for internal trips (I-I) 26 Figure 5.2 Mode choice model structure for external trips (I-E and E-I) 27 Figure 5.3 Trip’s segments represented by TODTrips 32

Appendices Appendix A TODTrips Trip Generation Model Land Use and Trip Rates (Source: SMC 2010 and RTA 2002) Appendix B TODTrips Trip Distribution Model and Parameters Appendix C TODTrips Mode Choice Parameters Appendix D Speed values used for different transport modes in TODTrips Appendix E Other model parameters used in TODTrips Appendix F Public Transport fare rates

Page iv 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

1. Introduction

1.1 Background

As part of the Stirling City Centre (SCC) - Light Rail Feasibility – Phase 2 study, Parsons Brinckerhoff (PB) conducted a desktop assessment of five alternative modal and operating environments options that could be available to serve the SCC. The range of options starts from bus on street, street car (tram on street), bus and light rail in exclusive lanes with priority (BRT and LRT) and light rail on single side (LRT sided).

This report describes the development of the public transport model using the TOD Trips package to estimate number of passenger trips for the 2031 proposed land use plan and ridership of the five alternative modes. Five alternative modal scenarios are:

 Base case with 2031 bus option

 Street Car

 LRT

 BRT

 LRT (Single sided).

The section of the Northern Railway Line to Joondalup between Stirling and Glendalough stations, local route 413 bus service, Circle Bus route 98/99 and route 400 were included in every scenario.

1.2 TODTrips model

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 1

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

The following outlines the broad capabilities and limitations of the TODTrips model:

Capabilities

 detailed modelling of land use patterns, including the distribution of population and employment inside the study area

 sophisticated generalised cost mode choice model for four main modes of travel – car, public transport, cycle and walk

 modelling of daily trips for all purposes that start and/or end in the study area.

Limitations

 coarse representation of zones outside the study area – modelling of links to key origins and destinations

 no modelling of detail road and traffic network – assumptions are made about car travel attributes

 no modelling of trips that start and end outside the study area – that is, no modelling of through trips.

1.3 Report outline

This report is presented in the following sections:

Section 2 – Model Scope – description of the overall TODTrips model

Section 3 – Land Use Modelling – description of the way land use patterns are modelled

Section 4 – Trip Generation and Distribution – description of the procedure for estimating the quantity and distribution of average daily trips

Section 5 - Mode Choice Modelling – details of the mode choice modelling component

Section 6 - Car use and public transport ridership for different scenarios.

Page 2 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

2. Model scope

2.1 Introduction

Key features associated with the TODTrips modelling platform applicable to the Light Rail Feasibility – Phase 2 study include:

 modelling of land use according to zoning types and floor space ratios

 modelling of higher density land use around selected transport nodes

 detailed modelling of bus and rail services together with consideration of alternative transit options including Street Car, Light Rail (LRT), BRT and LRT on single side

 modelling of public transport travel to key regional zones around study area

 estimation of daily trip patterns

 mode choice modelling based on generalised cost for car, public transport, cycling and walk trips.

The primary focus of TODTrips is the rapid development and testing of a range of scenarios related to land use and transport planning.

TODTrips model focused on the average daily person trips for all trip purposes for year 2031.

2.2 Study area and zone system

The study area consists of Stirling Centre and Osbourne Park as shown by the red dashed line in Figure 2.1. The development of TODTrips zones was based on detailed structure plan as well as STEM Travel Zones (TZs). The geographic scope of TODTrips encompasses the City of Stirling metropolitan region but with varying levels of aggregation – a finer zone system with 147 small zones was used within the study area with 33 key regional zones (broadly based on local government areas) covering the metropolitan area. Figure 2.2 depicts 147 internal zone system which was used for detailed modelling of land use and transport.

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 3

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 2.1 Stirling study area

Page 4 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 2.2 TODTrips zone system

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 5

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

2.3 Representation of network, services and access among zones

TODTrips models incorporate a full detailed representation of transit network and associated services within the study area. A number of TODTrips components were used to represent the transit network:

 GIS Point Objects: Components of this type represent the foundation of TODTrips spatial modelling toolbox as it provides the interface to the GIS and extract spatial data (i.e. x, y coordinates) to support the construction of network within TODTrips. GIS point objects were used to represent zone shapes (polygon) for zones within study area (internal zones), zone centroids (a single GIS point) for internal and external zones (outside study area) and transit stops (GIS point).

 Transit stops: Transit stops for any transit mode including bus, rail, street car, LRT and BRT are represented in detail within TODTrips. Key attributes of transit stops include Stop ID, Stop Name, Type of service (local and or regional services), and station with or without park-and-ride, transit modes, access fare, average platform access time and location with coordinates updated by GIS Points Objects.

 Lines and line segments: Every transit line is made up of a number of line segments. Each line segment is defined by two stops at the end points. Distance between these two end points defines the length of each segment. Given travel speed value, travel time on every segment can then be updated. Travel cost on every segment can also be updated with value of time of transit riders ($ per hour). Every transit line is associated with a number of public transport services so it can keep track of any change in service such as stopping pattern and time table.

 Transit services: Each transit service can be specified and associated with any particular transit line. Key service data includes Service ID, name, direction, stopping patterns, frequency of service and timetable data. If timetable data is not available then the travel time on any line segment can then be updated from travel speed.

For non transit modes of travel within study area, local access/egress and or external access/egress, straight line links were used to represent the following cases:

 Zone to zone access by non transit modes including car, walk and cycling modes: Straight line based centroid connectors were used to represent zone to zone access by non transit mode including car, walk and cycling. Different route directness factors were used to improve network representation for these main modes (see Appendices for model parameters and other assumptions).

 Local access and egress to and from zone to every transit stop by different access modes: A straight line link was used to connect origin zone centroids and every transit stop in the study area. For every origin zone TODTrips calculates the walk access distances to the closest transit stops. This feature offers more flexible and more detailed approach in examining the impact of specifying local access mode to the overall travel mode choice which normally includes local access to transit, in vehicle time using a particular transit mode and local egress. Detailed mode choice structure for internal to internal (I-I) movement within study area will be presented in more detailed in Section 4 of this report.

Page 6 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

 External access and egress to and from external zone to regional and or local transit stop by different access modes: A straight line link was used to connect origin external zone centroid and any specified transit stop in the study area. Different external access and/or egress modes and associated speeds can be specified, for example appropriate speeds for Regional Rail or Regional Bus services.

2.4 Public transport scenarios for 2031

TODTrips incorporates the following main modes of travel:

 car – driver and passenger combined

 walk and cycling combined

 rail (access via Glendalough and Stirling Stations within study area)

 bus (include local and regional bus services that pass through the study area)

 alternative transit options including bus, street car, LRT, BRT and LRT on single side.

Existing public transport services within the study area are described in Table 2.1 and Figure 2.1. The Northern rail service has Stirling and Glendalough stations as the key rail access points within study area. The speed and frequency of the Northern rail service are based on Transperth time table. The operating characteristics of bus services (including 98/99, 413 and 400) such as speed, number of stops and frequency of services were made available to the TODTrips study team by the PTA.

Table 2.1 Existing Public Transport’s operating environment for Stirling

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 4 3 413 Local bus 19.5 9 4 4 400 Regional bus 24.2 4 4

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 7

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 2.3 Stop locations of existing public transport service

These existing public transport services were included in every scenario and operating environment which will be described in the following sections

2.4.1 Scenario 1 – Base case (Bus)

Table 2.2 Scenario base (Bus) operating environment

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 4 3 413 Local bus 19.5 9 4 4 400 Regional bus 24.2 4 4 5 Base Local bus 15.0 13 4

Apart from the existing public transport services (Joondalup rail, 98/99, 413 and 400 bus services), new service is highlighted in Table 2.2 as Service number 5 and labelled as Base. In this scenario, a local bus service is tested with 13 stops (see Figure 2.4), average speed of 15 kph and run every 15 minutes.

Page 8 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 2.4 Stop locations of new local bus service in Scenario 1

2.4.2 Scenario 2 – Street Car

Apart from the existing public transport services (Joondalup rail, 98/99, 413 and 400 bus services), a new service is highlighted in Table 2.3 as Service number 5 and labelled as S2. In this scenario, a street car (or tram) is tested with 19 stops (see Figure 2.5), average speed of 15 kph and run every 5 minutes. It should be noted that the operating speed of service 400 is reduced from 24.2 kph down to 15 kph mainly due to safety as both the street car and service 400 could share the road space on the Scarborough Beach Road and general traffic including bus service 400 would have to give way to passengers alighting and boarding at tram stops.

Table 2.3 Scenario S2 (Street Car) operating environment

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 4 3 413 Local bus 19.5 9 4 4 400 Regional bus 15.0 4 4 5 S2 Street car 15.0 19 12

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 9

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 2.5 Stop location of street car service in Scenario 2

2.4.3 Scenario 3A – LRT

Apart from the existing public transport services (Joondalup rail, 98/99, 413 and 400 bus services), a new service is highlighted in Table 2.4 as Service number 5 and labelled as S3A. In this scenario, a dedicated LRT is tested with 11 stops (see Figure 2.6), average speed of 20 kph and run every 5 minutes. It should be noted that the operating speed of service 400 is reduced from 24.2 kph down to 20 kph mainly due to safety as both the LRT and service 400 would share the road space on the Scarborough Beach Road. However, the planned speed for the bus service 400 is still higher than in Scenario S2 with Street Car as passenger alighting and boarding at LRT stops will be more protected with LRT.

Table 2.4 Scenario 3A (LRT) operating environment

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 3 3 413 Local bus 19.5 9 9 4 400 Regional bus 20.0 4 4 5 S3A LRT 20.0 11 12

Page 10 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 2.6 Stop locations of LRT service in Scenario 3A

2.4.4 Scenario 3B – BRT

The operating environment of this scenario is identical to Scenario 3A in terms of existing public transport services and the alignment and associated stopping patterns of the new mode (see Figure 2.4). The only difference is that the new mode is changed from LRT to BRT and the frequency of service is reduced from 12 to 4 services per hour.

Table 2.5 Scenario 3B (BRT) operating environment

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 3 3 413 Local bus 19.5 9 9 4 400 Regional bus 20.0 4 4 5 S3B BRT 20.0 11 4

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 11

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 2.7 Stop locations of BRT service in Scenario 3B

2.4.5 Scenario 4 – LRT (single sided)

This scenario was constructed as a variation to Scenario S3A which is also a LRT option. However, the LRT proposed in this scenario is only operating on the southern side of Scarborough Beach Road (western side on Stephenson Boulevard). This mode of operation might create some distance constraints for residents on the northern side of Scarborough Beach Road as they might need to walk extra distance to pedestrian crossing to be able to cross Scarborough Beach Road. This extra walking distance (currently assumed to be 30 metres) was added to the utility function for those zones affected by this scenario.

Table 2.6 Scenario 4 (LRT sided) operating environment

Service Line ID Mode Speed (km/h) Stops Frequency 1 Joondalup Rail 58.5 2 3 2 98/99 Regional bus 24.2 3 3 3 413 Local bus 19.5 9 9 4 400 Regional bus 15.0 4 4 5 S4 LRT Side 17.5 16 12

Page 12 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 2.8 Stop locations of LRT (sided) service in Scenario 4

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 13

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

3. Land use considerations

3.1 Overview

TODTrips model incorporates a detailed modelling of land use development proposals/assumptions within the study area for each internal zone. The purpose of the land use modelling component of the model is to comprehensively represent the proposed future zoning of the area and land use intensities in order to estimate the quantum and distribution of future population and employment. By this method the impacts of increased densities in certain areas and changes in designated use are captured within the model.

3.2 Study area land use

3.2.1 Methodology

Figure 2.2 above shows the zoning system used in the study area. A total of 147 model zones were used to represent the 2031 future land use zoning for the Stirling study area. Gross Floor Areas (GFAs) and number of dwellings and dwelling types for each zone were specified as main input to the land use modelling module of TODTrips. These input data together with the rate of household size per dwelling and GFA (in m2) per employee for different types of developments were used to determine population and employment figures for 2031 base case.

The identification of relevant references and selection of suitable rates for estimating population and employment levels for different GFAs by different land use categories was an important task in the methodology. This procedure is described in detail in the following section.

3.2.2 2031 land use development

There are a number of recent studies (SKM in 2010 and Syme Marmion (SMC) in 2009) which focused for two key areas of the study area. As indicated on Figure 3.1, these two areas are Stirling Centre (labelled as precincts 1–7) and Osborne Park (labelled as precincts 8 and 9). Land use planning for these two areas and their contribution to the development of the whole study area in 2031 can be viewed by the distribution of GFAs across residential, retail, office, commercial, industrial and education land uses (see Table 3.1).

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 15

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 3.1 Study area by groups of precincts

Table 3.1 Distribution of GFAs across different land uses for Stirling study area in 2031

Residential Retail Office Commercial Industrial Education GFA GFA GFA GFA GFA Dwellings (m2) (m2) (m2) (m2) (m2) Precincts 1 to 7 13049 177747 322505 139498 175000 30000 Precincts 8 & 9 5500 70000 375000 650000 900000 0 (Osborne Park) Total (study area) 18549 247747 697505 789498 1075000 30000

Page 16 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

This distribution of GFAs across different land uses for Osborne Park are similar to the SMC mid range figures documented in its report (Table 19 – Projected Centre Parameter Values as at 2031) except that an extra 650,000 m2 was added to the commercial land use category following consultation with the Stirling Alliance team.

The household size per dwelling from the 2002 RTA New South Wales’s Guide to Traffic Generating Developments was adopted to estimate population figure for the Stirling study area (see Table 3.2).

Table 3.2 Average household size for different types of dwellings

Dwelling type Average household size Houses 3.0 Townhouses 2.5 Apartments 1.5 (Source: RTA NSW (2002) Guide to Traffic Generating Developments)

The selection of suitable rates for estimating employment figures for different GFAs was based on two references. The first reference was the RTA New South Wales’s Guide to Traffic Generating Developments (2002). The second source which was a study carried out by Syme Marmion & Co (SMC) in 2009 on the Scarborough Beach Road Population and Land Use Study. Table 3.3 compiles the floor space (in m2) per employee which are used in RTA and SMC reports.

Table 3.3 Floor space in m2 per employee from Syme Marmion & Co report and RTA NSW Guideline

From Syme Marmion Report (2009) From RTA NSW (2002) Shop/Retail 30 NA Other Retail 63 NA Retail (Average) 45 10.53(*) Office (Average) 25.5 21.05 Commercial (Average) NA 21.05 Industrial (Average) 114 21.05 Education (Average) NA 21.05 Note: (*) Retail average value for floor space (in m2) per employee was not available from RTA and was assumed in the initial calculation.

Comparing the floor space (in m2) per employee in Table 3.3 reveals that there are marked differences among the two references. The RTA NSW values are much lower than the values reported by Syme Marmion & Co Study. The Syme Marmion & Co (SMC)’s values of floor space per employees was adopted as the study was more recent (2009) than the RTA NSW (2002) and likely to be more relevant to the study area.

The floor space ratios per employee from the two sources were calculated in Table 3.4. These ratios were used to adjust the values used in 2002 RTA Guideline for average number of employees per 100 m2 GFA.

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 17

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Table 3.4 Adjusted average number of employees per 100 m2 GFA based on SMC and RTA NSW’s guideline

Average number of employees per 100m2 GFA Type of land use SMC equivalent rate RTA rate Adjusted RTA rate Retail (Average) 2.2 9.50 2.2 Office (Average) 3.9 4.75 3.9 Commercial (Average) NA 4.75 3.9 Industrial (Average) 0.9 4.75 0.9 Education (Average) NA 4.75 3.9

Given the adopted rate of household size per dwelling and GFA (in m2) per employee for different types of developments, the population and employment for Stirling study area in 2031 can be calculated as shown in detailed in Table 3.5 for Stirling Centre and Osborne Park across different land uses.

Table 3.5 Population and employment estimates for different land use categories in Stirling study area in 2031

Residential Retail Office Commercial Industrial Educational

) ) ) ) ) ) 2 2 2 2 2

A (m Dwellings Population GFA(m Jobs GF Jobs GFA(m Jobs GFA(m Jobs GFA(m Jobs

Precincts 1 to 7 13049 19574 177747 3910 322505 12578 139498 5440 175000 1575 30000 1170 Precincts 8 & 9 5500 8250 70000 1540 375000 14625 650000 25350 900000 8100 0 0 (Osborne Park) Total 18549 27824 247747 5450 697505 27203 789498 30790 1075000 9675 30000 1170

Table 3.6 provides a summary table of population and employment for Stirling Centre and Osborne Park. A total number of around 74,300 jobs were estimated for Stirling study area with Osborne Park contributing two thirds of employment (49,600 jobs) and Stirling Centre, one third (24,700 jobs). In contrast to employment figures, Stirling Centre contributes two thirds of population level (19,600 people) and the remaining third in Osborne Park) 8,300 people).

Table 3.6 Population and employment estimates for Stirling Study Area in 2031

Population Employment Precincts 1 to 7 19574 24673 Precincts 8 & 9 (Osborne Park) 8250 49615 Total 27824 74288

Page 18 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

3.3 External zones

A total of 33 external zones was established to represent travel movement from within the study area to outside the Stirling study area (denoted as I-E for Internal to External movement) and from outside travel to the study area (E-I for External to Internal movement). These 33 external zones consist of 31 LGAs and Stirling LGA remainder. The Stirling LGA remainder was represented by 2 external zones and labelled as Stirling Remainder East and Stirling Remainder West. Apart from car mode, the availability of transport access to and from these external zones to the study area by either regional bus and or rail service was specified in the network input to TODTrips (see Table 3.7).

Section 4 of this report describes how travel between the study area and these external zones is modelled within TODTrips.

Table 3.7 List of 33 external zones of TODTrips model for Stirling Study

Number LGA Regional rail access Regional bus access 1 Wanneroo True False 2 Joondalup True False 3 Stirling Remainder West False True 4 Stirling Remainder East False True 5 Cambridge False True 6 Subiaco False True (*) 7 Nedlands True False 8 Claremont True False 9 Cottesloe True True 10 Peppermint Grove True True 11 Mosman Park True True 12 Perth True True 13 Vincent False True 14 Bayswater True False 15 Bassendean True False 16 Swan True False 17 Mundaring True False 18 Kalamunda True False 19 Belmont True True 20 Victoria Park True True 21 South Perth True False 22 Melville True False 23 East Fremantle True True 24 Fremantle True True 25 Cockburn True False 26 Canning True False 27 Gosnells True False 28 Armadale True False 29 Serpentine-Jarrahdale True False

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 19

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Number LGA Regional rail access Regional bus access 30 Kwinana True False 31 Rockingham True False 32 Murray True False 33 Mandurah True False Note: (*) Assume a service providing direct connection will be available in 2031

Page 20 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

4. Trip generation and distribution

4.1 Overview

The TODTrips model focuses on the average weekday all trip purpose trips. Estimates are made of person trip productions and attractions for each zone inside the study area based on the land use development input for 2031 as described in Section 3. These person trips are then distributed internally as well as across the external zones. The model does not include trips that start and end outside the study area – thus, any trips that may pass through the study area are not included.

4.2 Trip generation

Estimates of average daily person trip productions and attractions were made within TODTrips for each study area zone using trip rates applied to population for residential centres and employment GFAs for employment centres. For residential zones, the average week day trip rate per person value of 3.75 trips per person is adopted from the RTA New South Wales’s Guide to Traffic Generating Developments (2002).

For employment centres, the selection of suitable trip rates for trip productions and attractions were based on two sources as described in Section 3. As discussed, there are marked differences in the floor space ratios per employee from the two sources. The adjusted average daily trips per 100 square metres based on the values from SMC and RTA’s GFA per employee from Table 3.3 above are presented in Table 4.1. These ratios were used to adjust the values used in 2002 RTA Guideline for average daily trip rate per 100 square metres.

Table 4.1 Adjusted Average Daily Trip per 100 m2 GFA based on SMC and RTA NSW’s Guideline

Average Daily Trip rates per 100 sqm GFA SMC and RTA’s GFA Adjusted Average Type of land use RTA Average Daily per employee ratio Daily Trip rate based Trip rate per 100 sqm on RTA and SMC Retail (Average) 4.29 86 20 Office (Average) 1.21 11 9 Commercial (Average) NA 11 9 Industrial (Average) 5.42 5.5 1 Education (Average) NA 11 9

Table 4.2 provides a detailed table of GFAs (in m2) and trip generation estimates for Stirling Centre and Osborne Park from residential, retail, office, commercial, industrial and education land use types.

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 21

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Table 4.2 Trip generation estimates for different land use categories in Stirling Study area in 2031

Residential Retail Office Commercial Industrial Educational ) ) ) ) ) )

2 2 2 2 2 FA(m Population Persontrips out in & G Persontrips out in & GFA (m Persontrips out in & GFA (m Persontrips out in & GFA (m Persontrips out in & GFA (m Persontrips out in &

Precincts 1 to 7 19574 73400 177747 35550 322505 29026 139498 12554 175000 1750 30000 2700 Precincts 8 & 9 8250 30929 70000 14000 375000 33750 650000 58500 900000 9000 0 0 (Osborne Park) Total 27824 104329 247747 49550 697505 62776 789498 71054 1075000 10750 30000 2700

Even though there are differences in land use types and the level of development for Stirling Centre and Osborne Park, the daily trip generation estimates for the two areas are contributing equally to the trip generated from the whole study area. As indicated in Table 4.3, Stirling Centre contributes 51% and Osborne Park contributes 49% to the total of around 300,000 person trips from the study area.

Table 4.3 Average Daily Trip Generation estimates for Stirling Study area in 2031

Total daily person trips in and out Percent Precincts 1 to 7 154981 51% Precincts 8 & 9 (Osborne Park) 146181 49% Total 301162 100%

4.3 Trip distribution

4.3.1 Overview

TODTrips model handles trip distribution differently to most traditional four-step models by using a combination of specified parameters and gravity model techniques. Trip distribution is done in this way to avoid problems that can arise from uncontrolled application of a gravity model to growth areas where new trip productions are sometimes distributed to a small number of local zones.

For this study, the modelling of future scenarios is based on changes in travel patterns within the study area, and to and from the study area. In order to support this modelling requirement, detailed representation of study area of 147 zones was maintained. In addition, the transport access and interaction between the 147 zones within study area (internal zones) and external 33 LGAs are also modelled within TODTrips. Figure 4.1 describes modelling scope and assumptions used in distributing trips generated from the study area.

Page 22 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 4.1 Trip distribution modelling – scope and assumption used for Stirling Study

The objective of the trip distribution module is to distribute trips across the following:

 Internal trips – those inter-zonal trips that start and end inside the study area (labelled as I-I in Figure 4.1) and

 External trips – those inter-zonal that pass between internal and external zones (labelled as I-E and E-I in Figure 4.1).

4.3.2 Internal trips

The total amount of internal trips within the study area is assumed to be 30.5% of total number of trips generated by the study area. This self containment figure of 30.5% is based on the output from STEM model for year 2031. Given a total number of around 300,000 trips generated from the study area, the total amount of internal trips within the study area is equal to some 92,000 daily trips (=30.5% x 301162 trips). This total trip figure of 92,000 daily trips together with the zonal trip production and trip attraction are the key input to the distribution model (applying a gravity model). A detailed description of the gravity model is presented in Appendix B.

4.3.3 External trips

The output from 2031 STEM model indicates that the total amount of internal trips within study area is estimated to be 30.5% and remaining 69.5% of total number of daily trips generated from the study area consists of trips that have either origin or destination from outside study area. Table 4.4 provides detailed distribution of approximately 300,000 daily trips generated from the study area among internal zones and in relation to the external zones.

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 23

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Table 4.4 Distribution of Average Daily Trips generated from Study Area in 2031

Destination Total daily trips Internal Origin Origin produced from study (147 zones) area I-I matrix I-E matrix 150581 Internal Dimension = 147 x 147 Dimension = 147x33 =46026 (I-I trips) (147 zones) Daily trips in = 46026 Daily trips out = Plus Daily trips out = 46026 104555 104555 (I-E trips) E-I matrix External Not included in the Dimension = 33 x 147 (33 zones) model Daily trips in = 104555 150581 Total daily trips to and Total daily trips =46026 (I-I trips) from study attracted into study Plus 301162 area 104555 (E-I trips) (= 150581 + 150581)

Page 24 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

5. Mode choice model

5.1 Model structure

Two mode choice models were developed for estimating the mode choice of internal trips (I-I) and external trips, respectively. Description of these different two mode choice models are presented in the following sections.

5.1.1 Mode choice model for internal trips (I-I)

Figure 5-1 describes the structure of the mode choice model used for representing the internal trips (I-I) within Stirling study area. The model structure is a nested logit model with two levels. At the top level, a total of three travel choices are considered to be available. They are car, cycle and walk and transit. The car mode includes the use of any private motorised vehicle as driver or passenger. Cycle and walk trips are those trips where these are the only modes used (that is, a walk / bus trip is classified as a bus trip). Transit choice set considers all public transport modes that are available to a specific scenario. As described section 2 above, rail choice (i.e. Northern line between Stirling and Glendalough stations) and existing bus choices (include services 98/99, 400 and 413) are present in the choice set of public transport for every scenario. Available alternative transport modes specified are 2031 Bus (Base scenario), Street car (scenario S2), LRT (scenario S3A), BRT (scenario S3B) and LRT one-sided (scenario 4).

In general, the structure is intended to replicate expected mode choice behaviour in that it assumes people first consider whether to use car, walk and cycle or transit (higher level choice set) and then consider specific public transport mode within transit choice set which includes rail, bus and alternative mode such as street car, LRT, or BRT. Walk is modelled as the access mode to transit stations.

At each level of the nested model, the split among alternatives is done as a multinomial logit based on estimated generalised costs of the alternatives. At higher levels, the composite generalised costs representing transit choice is calculated as the log sum of utility in the transit choice set. The following formulas applied to all alternative choices at the top and at the transit level.

For every OD (origin destination) pair, N probabilities pi are calculated to represent travel’s

preferences among N choices. In the formulae below, pi represents the probability of selecting choice i to travel from origin O to destination D. For simplicity, the two subscripts O and D are dropped in the formulae. exp() is a exponential function and GCi is generalised cost for choosing choice i to travel between OD. It is a linear function of the set of observed

travel attributes that influence choice i. The sum of pi is equal to 1.

GCi )exp( pi = =Nj ∑ GC j )exp( j=1 At each level of the nested model, the split among alternatives is done using the logit formula

pi as a multinomial logit based on estimated generalised costs of the alternatives. At higher levels, the composite generalised costs representing transit choice is calculated as the log sum of utility in the transit choice set.

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 25

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 5.1 Mode choice model structure for internal trips (I-I)

5.1.2 Mode choice model for external trips (I-E and E-I movements)

Figure 5.2 describes the structure of the mode choice model used for representing the external trips to and from study area (E-I and I-E movements). The model structure is similar to the structure of mode choice for internal trips as it is also a nested logit model with two levels. However, the difference is at the top level where car, regional rail and regional bus are the three travel choices are available. This model structure is used for representing both E-I and I-E movements as follows:

 Car was considered as the only main mode with direct access between external zones and internal zone.

 Regional rail and regional bus were modelled as the two transit modes providing external access (for E-I movements) or external egress (for I-E movements) to connect external zones to regional stations located within study area. Regional rail stations are Stirling and Glendalough stations. Regional bus stations include stops of service 400 and service 98/99.

 Transfer by walk mode represents the travel segments connecting regional rail and regional bus stops to the nearest stops of local transit network.

Page 26 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

 Transit choice set considers all local public transport modes that are available to a specific scenario. As described Section 2 above, rail choice (i.e. Joondalup line between Stirling and Glendalough stations) and existing bus choices (include services 98/99, 400 and 413) are present in the choice set of public transport for every scenario. Available alternative transport modes specified are 2031 Bus (Base scenario), Street Car (Scenario S2), LRT (Scenario S3A), BRT (Scenario S3B) and LRT sided (Scenario 4).

 Walk represents the local access mode (for I-E movements) or local egress mode (for E-I movements) to and from local transit network of rail, bus, street car, LRT and BRT.

Figure 5.2 Mode choice model structure for external trips (I-E and E-I)

5.2 Travel attributes

5.2.1 Overview

The mode choice model is based upon generalised cost estimates for alternative modes which, in turn, are based upon estimates of travel attributes. The PT model uses the following processes for estimating travel attributes for different modes:

 car – travel attributes including travel times and costs are specified outside the model (that is, there is no modelling of the traffic network)

 public transport – travel attributes, including travel time, walk access/egress time, wait time, transfers and fares, are estimated within the PT model based on specified public transport routes and services

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 27

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

 walking and cycling – walk and cycle travel times are estimated based on specified average speeds and distance factors.

Generalised costs for travel between each zone pair are built separately for each of these three modes based on the estimated travel attributes.

Table 5.1 below provides a summary of the travel attributes for each mode in the Stirling study.

Table 5.1 Attributes of main travel mode used in the Stirling Study

Travel mode Attributes Car operating costs Car In-vehicle time Other costs (parking, toll) Walk or cycle time Walk & cycle Walk or cycle long distance constraints Total fare In-vehicle time Access time Transit Wait time Transfer time Egress time

Appendices D to F provide detailed values and assumption used for the travel attributes in the Stirling Study.

5.2.2 Car travel

The following attributes of car travel are specified for internal and external trips:

 distance factor – applied to straight line distance between zones to allow for the density of the road network

 average speed – specified separately for internal travel and to and from each external zone

 car operating costs – average operating cost in $ per km

 parking costs – destination parking cost, $ per trip

 parking time – destination time for parking including car parking time and walking time.

The assumed vehicle operating cost is $0.50 per km based on the ATC Guidelines (Volume 5, p. 42) with an adjustment to reflect 2010 prices.

Page 28 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

5.2.3 Public transport travel

Travel by public transport services (bus or rail) are modelled as a combination of access, main mode and egress segments.

Walk access/egress to transit stops was modelled for all zones in the study area.

Travel attributes for public transport travel are estimated by the TODTrips model based on specified public transport networks and services. Transit services running on the network are specified separately for each scenario (see Section 3).

Public transport service attributes include service frequency and travel times between stations and stops – the latter is based on either specified schedule times or network speeds. Public transport travel between the study area and external zones is modelled by way of specifying key regional bus and rail services.

The TODTrips model applies shortest path algorithms to select a set of shortest paths between each origin and destination zone pair based on the specified network and services. Walk access to different adjacent parts of the public transport network is modelled from the centroid of each zone with parameters applied to allow for the degree of permeability in each zone.

Once the shortest public transport paths between each zone pair are determined, the following public transport travel attributes are estimated:

 walk access and egress time

 total service waiting times

 total service in-vehicle times

 transfer times

 fare (see Appendix F).

5.2.4 Walking and cycling

Walking and cycling times for travel between each origin and destination zone pair are estimated from applying specified distance factors and average speeds to the straight line distances between zones. The distance factors are used to allow for different densities in the walking and cycling network. Cycling average speeds can be varied to allow for the effect of cycle facilities.

For the Existing Base Case (2001) scenario the following parameters were used:

 average walk speed = 4.0 km/h

 average cycle speed = 15 km/h.

Constraints were applied as penalty to the walk and cycle mode for longer walking and cycling distances (see Appendix E).

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 29

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

5.3 Generalised costs of travel

5.3.1 Formulating generalised costs to represent different trip segments between origin and destination

TODTrips model maps all key segments of trips between an origin and a destination in a multi-modal transport network context. Figure 5.3 presents trip’s segments between origin zone and destination zone. These trips are mapped into a complete mode choice structure so that the cost of travel by different transport mode choices can be identified, formulate and calculated. In TODTrips, generalised costs are converted to generalised time (labelled as GT) in minute unit.

 Car, walk and cycle modes provide a direct link between origin zone and destination zone so there is no access and egress components. The generalised times of car, walk and cycle modes are calculated by the following formulas:

GT by Car = GC weight main mode x Travel time by Car + GT vehicle operating cost by Car + GT other cost by Car

GT by Walk = GC weight main mode x Travel time by Walk + GT penalty long distance by Walk

GT by Cycle = GC weight main mode x Travel time by Cycle + GT penalty long distance by Cycle

Where: Travel time by Car = ((Distance x Car_RouteDirectnessFactor)/(Car_speed*1000) * 60) (in minutes) Travel time by Walk = ((Distance x Walk_RouteDirectnessFactor)/(Walk_speed*1000) * 60) (in minutes) Travel time by Cycle = ((Distance x Cycle_RouteDirectnessFactor)/(Cycle_speed*1000) * 60) (in minutes) GTvehicle operating cost by Car = ((VehOpercarCost/1000 * distance)/car_VOT) * 60 (in minutes) GT of other cost = (parking cost + toll cost)/car_VOT * 60 (in minutes) Distance = straight line distance between origin and destination zone centroids (in metres) Car_speed = average car speed (in kph) Walk_speed = average walk speed (in kph) Cycle_speed = average cycle speed (in kph) Car_VOT=value of time of Car driver and passenger (in $/hour) VehOpercarCost = car operating cost (in $/km) Car_RouteDirectnessFactor, Walk_RouteDirectnessFactor and Cycle_RouteDirectnessFactor = factors to convert straight line distance between origin and destination to network distance. GC weight main mode = weight value applied to main mode.

The specific values of generalised time of car, walk and cycle modes for every origin and destination are based on the distance values which are calculated by TODTrips given the values of other parameters. Lists of parameter values adopted for the generalised time calculation for Stirling study is in the Appendices C to F.

 Transit involves three trip segments: access, main and egress. Transfer is also considered to allow for transfer by walk across different stations/stops. Transfer segment is particularly important in handling external trips where the transfers at regional stations (such as Stirling and Glendalough stations) and regional bus stops (service 98/99 and service 400) are required to connect to the local transit network.

 Local access and local egress within Stirling study area is by walk mode. External access/egress to and from external zones (i.e. LGAs) is by regional rail and or regional bus services.

Page 30 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

 Main transit modes include rail, bus, and alternative modes introduced by each specific scenario (2031 bus, street car, LRT, BRT and LRT single sided).

 The generalised times of transit modes for internal trips (I-I movements) are calculated by the following formulas:

GT by transit mode = GC weight access mode x (Travel time by Walk + GT equivalent of transit fare + GC weight main mode x GT In vehicle travel time by transit + GC weight wait x waiting time + GC weight egress mode x GT egress cost by Walk

Where: Travel time by Walk = ((Distance x Walk_RouteDirectnessFactor)/ (Walk_speed*1000) * 60) (in minutes) GT equivalent of transit fare = (Fare/Transit_VOT) * 60 (in minutes) Transit_VOT=value of time of transit riders (in $/hour) Walk_speed = average walk speed (in kph) GT In vehicle travel time by a transit mode = (Distance/(speed of transit mode x 1000) )* 60 (in minutes) Distance = straight line distance between origin and destination stations/stops (in metres) Speed = speed of a specific transit service (in kph) Average waiting time = average waiting time at transit stops (in minutes). Waiting time was assumed to be equal to half of the average headway of a particular transit service in every scenario. GC weight access mode, main mode and egress, = weight values applied to access, main and egress modes, respectively.

The specific values of generalised time of transit modes for every origin and destination are based on the distance values which are calculated by TODTrips given the values of other parameters. Lists of parameter values adopted for the generalised time calculation for Stirling study is in Appendices C to F.

 The generalised times of transit modes for external trips, particularly I-E movements, are calculated by the following formulas:

Where: Travel time by Walk = ((Distance x Walk_RouteDirectnessFactor)/ (Walk_speed*1000) * 60) (in minutes) GT equivalent of transit fare = (Fare/Transit_VOT) * 60 (in minutes) Transit_VOT=value of time of transit riders (in $/hour) Walk_speed = average walk speed (in kph) GT In vehicle travel time by a transit mode = (Distance/(speed of transit mode x 1000) )* 60 (in minutes) Distance = straight line distance between origin and destination stations/stops (in metres) Speed = speed of a specific transit service (in kph) Average waiting time = average waiting time at transit stops (in minutes). Waiting time was assumed to be equal to half of the average headway of a particular transit service in every scenario. Transfer time by Walk = ((Transfer distance x Walk_RouteDirectnessFactor)/(Walk_speed*1000) * 60) (in minutes) GT equivalent of transit fare = (Fare/Transit_VOT) * 60 (in minutes) GC weight access mode, main mode, transfer and egress, = weight values applied to access, main, transfer and egress modes, respectively.

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 31

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Figure 5.3 Trip’s segments represented by TODTrips

5.3.2 Weighting trip segments used in calculating generalised costs of travel

The travel attributes for travel by each alternative mode are combined into a single generalised cost for that mode using weights. The different values for the weights allows for different valuations of the components of travel (compared with in-vehicle travel time). For example, a weight of 1.8 is applied to transfer time to allow for the fact that most people value transfer time at a rate about 80 % higher than in-vehicle travel time.

The values of generalised cost weights can be derived from stated preference surveys targeted at the particular markets of interest. In the absence of such survey data, values can be specified based on established guidelines and interpretation for the local context.

For the Stirling model, the Australian Transport Council guidelines were used to specify appropriate values for the weights for car and public transport travel – these are summarised in Table 5.3 below. More details of mode choice parameters used are in Appendices C to F.

Page 32 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Table 5.2 Generalised cost weights

Mode Name Value Access GC_weight_access 1.5 Main GC_weight_main 1.0 Egress GC_weight_egress 1.5 Transfer GC_weight_transfer 3.0

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 33

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

6. Assessment result of five transit modal scenarios

The assessment of five scenarios described in Section 3 was implemented with two car use scenarios: high and low car use for internal trips and high car use for external trips. Results of TODTrips model runs output for all scenarios are presented and discussed in the following sections.

6.1 Internal trips – mode share and ridership estimates

This section presents mode share and ridership estimates for all five scenarios to accommodate approximately 92,000 daily person trips estimated for internal travel movements (I-I). Tables 6.1 and 6.2 present the mode share results based on the high and low car scenarios. Main findings are as follows:

 Walk and cycle mode share is quite consistent and stable across different scenarios with average of 26.8% and 27.9% share in low and high car scenarios, respectively.

 Car, rail and bus had highest share in base scenario among all five scenarios.

 With the introduction of alternative transit modes including Street car, LRT, BRT and LRT single sided, mode choice pattern was redistributed where alternative modes gain an average share of 20%.

 These gains came from the drops in car share by around 7%, walk and cycle mode share by around 2%, rail share by around 3.5% and bus share by 3 to 8%.

Table 6.1 Low car use – Mode share for internal trips

by by by by by by by Scenario Car Walk & Cycle Rail Street Car LRT BRT Bus Base 47.3% 28.0% 8.3% 0.0% 0.0% 0.0% 16.4% Street Car 40.6% 26.4% 3.2% 24.5% 0.0% 0.0% 5.2% LRT 40.8% 26.5% 3.1% 0.0% 22.3% 0.0% 7.3% BRT 41.1% 26.6% 3.6% 0.0% 0.0% 20.5% 8.2% LRT (single sided) 40.5% 26.4% 3.1% 0.0% 24.0% 0.0% 5.9%

Table 6.2 High car use scenario – Mode share of internal trips

by by by by by by by Scenario Car Walk & Cycle Rail Street Car LRT BRT Bus Base 54.9% 29.6% 6.6% 0.0% 0.0% 0.0% 8.9% Street car 45.3% 27.4% 2.8% 20.4% 0.0% 0.0% 4.2% LRT 45.6% 27.5% 2.7% 0.0% 18.6% 0.0% 5.7% BRT 46.2% 27.7% 3.0% 0.0% 0.0% 16.7% 6.4% LRT (single sided) 45.3% 27.5% 2.7% 0.0% 19.9% 0.0% 4.7%

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 35

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Tables 6.3 and 6.4 provide a summary of ridership share of 92,000 daily person trips among different transport modes for low and high car use. Main findings are as follows:

Table 6.3 Low car use scenario – ridership share among different transport modes for internal trips

by by by by by by by Scenario Car Walk & Cycle Rail Street Car LRT BRT Bus Base 43575 25756 7626 0 0 0 15095 Street car 37331 24342 2965 22588 0 0 4826 LRT 37543 24399 2847 0 20563 0 6699 BRT 37816 24457 3300 0 0 18891 7588 LRT (single sided) 37327 24345 2874 0 22101 0 5406

Table 6.4 High car use scenario – ridership share among different transport modes for internal trips

by by by by by by by Scenario Car Walk & Cycle Rail Street Car LRT BRT Bus Base 50575 27233 6096 0 0 0 8148 Street car 41667 25260 2546 18744 0 0 3835 LRT 41976 25343 2442 0 17080 0 5212 BRT 42493 25453 2807 0 0 15408 5892 LRT (single sided) 41695 25271 2454 0 18277 0 4355

6.2 External trips – mode share and ridership estimates

This section presents mode share and ridership estimates for all five scenarios to accommodate some 210,000 daily person trips estimated for external travel movements (I- E and E-I movements). Tables 6.5 present the result for the distribution of regional access and egress by external trips into and out of the study area. Main findings are as follows:

 Among a total of 210,000 trips (two way travel), mode share by car was estimated at 65% with 135,000 daily person trips and mode share by regional transit services share the remaining 35% with regional rail share value of 22% and regional bus share value of 13%.

 The introduction of alternative transit modes including Street Car, LRT, BRT and LRT single sided did make some impact (even not significant) on car mode share. The drop in car share by around 3% is noted. This could be due to the fact that the alternative transit modes connect regional transit services and final destination within study area.

Page 36 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Table 6.5 Distribution of regional access and egress by external trips into and out of study area (I-E and E-I movements)

Among the 210,000 daily person trips estimated for external travel movements (I-E and E-I movements), an average of 130,000 car trips had direct access between external and internal zones. While 130,000 daily person trips of car mode to and from external zones will become internal travel component within the study area, the 80,000 (=210,000-130000) daily person trips by regional rail and or regional bus will be connected to the local transit network within the study area. Tables 6.6 and 6.7 present the result for the modal split and ridership share by local transit modes from these approximate 80,000 daily person trips in connecting to the local transit services. Main findings are as follows:

 In Base scenario, rail and bus mode shares are 27.9% and 7.1%, respectively.

 Ridership estimates for alternative transport modes ranges from average of 26,000 daily person trips (BRT) to 34,000 daily person trips (street car, LRT and LRT single sided).

Table 6.6 Modal split of external trips using local transit services

by by by by by by Scenario Car Walk & Cycle Rail Street Car LRT BRT Base 50575 27233 6096 0 0 0 Street car 41667 25260 2546 18744 0 0 LRT 41976 25343 2442 0 17080 0 BRT 42493 25453 2807 0 0 15408 LRT (single sided) 41695 25271 2454 0 18277 0

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 37

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Table 6.7 Ridership estimates of external trips (I-E and E-I movements) using local transit services

by by by by by by Scenario Car Rail Street Car LRT BRT Bus Base 135900 58280 0 0 0 14931 Street car 131015 31901 34183 0 0 12011 LRT 129549 34097 0 32470 0 12995 BRT 130134 37668 0 0 26635 14674 LRT (single sided) 129623 33081 0 34176 0 12230

6.3 Combined internal and external trips - mode share and ridership estimates

This section presents mode share and ridership estimates for all five scenarios to accommodate a total of around 300,000 daily person trips estimated for all travel movements (including internal and external trips). Tables 6.8 and 6.9 present the mode share results based on the high and low car scenarios. Main findings are as follows:

 The 5% difference between BRT and street car and LRT is mainly due to the frequency of service of BRT is 15 minutes versus 5 minutes.

 Street car mode gains highest mode share is mainly due to its service coverage with 19 stops in comparing to 13 stops in LRT scenario.

 In terms of ridership estimates, all alternative modes scenarios are comparable and their values are in the range of 45,000 to 55,000 person trips per day.

Table 6.8 Low car use – Mode share for combined internal and external trips

by by by by by by Scenario by Rail Car Walk & Cycle Street Car LRT BRT Bus Base 59.6% 8.6% 21.9% 0.0% 0.0% 0.0% 10.0% Street car 55.9% 8.1% 11.6% 18.9% 0.0% 0.0% 5.6% LRT 55.5% 8.1% 12.3% 0.0% 17.6% 0.0% 6.5% BRT 55.8% 8.1% 13.6% 0.0% 0.0% 15.1% 7.4% LRT (single sided) 55.4% 8.1% 11.9% 0.0% 18.7% 0.0% 5.9%

Page 38 2106689A-PR_2812 PARSONS BRINCKERHOFF

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

Table 6.9 High car use – Mode share for combined internal and external trips

by by by by by by by Scenario Car Walk & Cycle Rail Street Car LRT BRT Bus Base 61.9% 9.0% 21.4% 0.0% 0.0% 0.0% 7.7% Street car 57.3% 8.4% 11.4% 17.6% 0.0% 0.0% 5.3% LRT 57.0% 8.4% 12.1% 0.0% 16.5% 0.0% 6.0% BRT 57.3% 8.5% 13.4% 0.0% 0.0% 14.0% 6.8% LRT (single sided) 56.9% 8.4% 11.8% 0.0% 17.4% 0.0% 5.5%

Table 6.10 Low car use – ridership estimates for combined internal and external trips

by by by by by by by Scenario Total Car Walk & Cycle Rail Street Car LRT BRT Bus Base 179474 25756 65905 0 0 0 30026 301162 Street car 168346 24342 34866 56771 0 0 16836 301162 LRT 167092 24399 36944 0 53033 0 19694 301162 BRT 167949 24457 40968 0 0 45526 22261 301162 LRT (single sided) 166950 24345 35955 0 56277 0 17635 301162

Table 6.11 High car use – ridership estimates for combined internal and external trips

by by by by by by by Scenario Total Car Walk & Cycle Rail Street Car LRT BRT Bus Base 186474 27233 64375 0 0 0 23079 301162 Street car 172682 25260 34447 52927 0 0 15845 301162 LRT 171524 25343 36539 0 49550 0 18206 301162 BRT 172627 25453 40475 0 0 42043 20565 301162 LRT (single sided) 171317 25271 35536 0 52453 0 16585 301162

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 39

Stirling City Centre - Light Rail Feasibility Study - Phase 2 TOD Trips Model Working Paper

7. Conclusions

The modelling results indicate that there is strong potential for the operation of an effective internal transit system in the Stirling Centre. The following conclusions and observations are made in relation to the results of the modelling:

 The transit system has a strong role to play in minimising the use of private motor vehicles for movement within the centre and minimising the demand for parking.

 With regards the modes tested, the Streetcar shows that it has the potential to attract marginally more passengers than the other options. This mode is the most accessible, with the highest number of stops which underlines the importance of selecting a mode which can be closely integrated with development along the corridor.

PARSONS BRINCKERHOFF 2106689A-PR_2812 Page 41

Appendix A

TODTrips Trip Generation Model Land Use and Trip Rates (Source: SMC 2010 and RTA 2002)

Table A-1: Land use and Trip Rates used for Stirling Study

Category Variables Values Population AvgWeekDayTripRateperPerson 3.75 Retail small AvgWeekDayTripRateper100smGFA 132 Retail medium AvgWeekDayTripRateper100smGFA 86 Retail large AvgWeekDayTripRateper100smGFA 55 Retail AvgWeekDayTripRateper100smGFA 20 Commercial AvgWeekDayTripRateper100smGFA 9 Office AvgWeekDayTripRateper100smGFA 9 Industrial AvgWeekDayTripRateper100smGFA 1 Education AvgWeekDayTripRateper100smGFA 9 Industrial AvgNumberWorkerper100smGFA 0.9 Office AvgNumberWorkerper100smGFA 3.9 Commercial AvgNumberWorkerper100smGFA 3.9 Retail AvgNumberWorkerper100smGFA 2.2 Education AvgNumberWorkerper100smGFA 3.9 Houses Density_in_dwg per ha 12 Townhouses Density_in_dwg per ha 20 Apartments Density_in_dwg per ha 40 Houses Average HHsize 3 Townhouses Average HHsize 2.5 Apartments Average HHsize 1.5 Retail FSR 1 Commercial FSR 3 Industrial FSR 0.4 Office FSR 3 Retail AverageGFA perworker 30 Commercial AverageGFA perworker 15 Industrial AverageGFA perworker 100 Office AverageGFA perworker 15

Appendix B

TODTrips Trip Distribution Model and Parameters

Estimating trip distribution using gravity model for internal trip movements (I-I) The gravity model, formulated as in the following equation, was used to estimate the pattern of travel within Stirling study area.

Tij = ki kj Ti Tj / f(tij) A-1

Where

 Tij is the estimate of the amount of trips between zone i and zone j;

 Ti and Tj are trip production from zone i and trip attraction to zone j;

 k is a proportionality factor; and

 f(tij) is a generalised function of the travel cost between zone i and zone j and it often receives the name of ‘friction function’ or ‘deterrence function’ because it represents the disincentive to travel as travel cost increases. For Stirling case study, f(tij) = tijalpha where alpha=1.2 and tij = distance in kilometres between centroids of origin zone i and destination zone j.

 ki and kj are calibration parameters and defined as follows:

-1 ki = {ΣkjTj / f(Tij)} A-2

-1 kj = {ΣkiTi / f(Tij)} A-3

An iterative process described below is necessary: given a set of values for the friction function, start with all kj = 1, solve for ki and then use these values to re-estimate the kj’s; repeat the estimated matrix satisfy the trip ends constraints.

 set all kj = 1.0 and solve for ki; in this context. ‘solve for ki’ means find the correction factors ki that satisfy the trip generation constraints (given from the output of trip generation module);

 with the latest ki solve for kj, e.g. satisfy the trip attraction constraints (given from the output of trip generation module);

 keeping the kj ’s fixed, solve for ki and repeat steps (2) and (3) until the changes are sufficiently small.

Output:

A full OD matrix of NxN, where N is total number of zones, would be produced. The numerical value in each matrix cell ODij represents the amount of person trips between origin zone i and destination zone j.

Appendix C

TODTrips Mode Choice Parameters

Table C-1: Weight values used in Mode Choice Models for Stirling Study

Mode Name Value Access GC_weight_access 1.5 Main GC_weight_main 1.0 Egress GC_weight_egress 1.5 Transfer GC_weight_transfer 3.0

Table C-2: Other weight values used in Mode Choice Models for Stirling Study

Model Parameters Name Value Internal Exponential function sensitive trip mode parameter for generalised cost choice of internal trips GC_SensParameter_Int 0.065 Internal Scale value applied to logsum trip mode of transit choices in low car use choice scenario GC_ScaleIntAccessUtilityforLowCarUse 1.50 Internal Scale value applied to logsum trip mode of transit choices in high car choice use scenario GC_ScaleIntAccessUtilityforHighCarUse 3.00 External Exponential function sensitive trip mode parameter for generalised cost choice of external trips GC_SensParameter_Ext 0.008 External trip mode Scale values applied to car choice choice utility GC_ScaleExtAccessUtility_forCar 2.00 External trip mode Scale values applied to regional choice rail's utility GC_ScaleExtAccessUtility_forRegionalRail 1.00 External trip mode choice Scale value applied regional model bus's utility GC_ScaleExtAccessUtility_forRegionalBus 1.00

Appendix D

Speed values used for different transport modes in TODTrips

Table D-1: Speed values used for different transport modes in Stirling Study

Average Transport modes Scenario Speed (kph) Car (local) All 35.0 Car (regional) All 55.0 Walk All 4.0 Cycle All 10.0 Rail All 58.5 Bus - Service 98/99 All 24.2 Bus - Service 413 All 19.5 Bus - Service 400 Base 24.2 Bus - Service 400 S2 15.0 Bus - Service 400 S3A 20.0 Bus - Service 400 S3B 20.0 Bus - Service 400 S4 15.0 2031 Bus Base 15.0 Street car S2 15.0 LRT S3A 20.0 BRT S3B 20.0 LRT (Single sided) S4 17.5

Appendix E

Other model parameters used in TODTrips

Table E-1: Other model parameters

Mode Variables Values Car Car operating cost per km $0.50 Car Fix car cost (parking, toll) $0.00 Car Value of time for car driver & passenger $6.80 Transit Value of time for public transport user $6.80 Car Route directness factor for car trip 1.2 Walk Route directness factor for walk trip 1.2 Cycle Route directness factor for cycle trip 1.4 Walk Travel time penalty for Walk from 500 - 1000 metres 8 minutes Walk Travel time penalty for Walk from 1000 - 1500 metres 16 minutes Walk Travel time penalty for Walk > 1500 metres 24 minutes Cycle Travel time penalty for Cycle from 1000 - 2000 metres 8 minutes Cycle Travel time penalty for Cycle from 2000 - 3000 metres 16 minutes Cycle Travel time penalty for Cycle from 3000 - 4000 metres 24 minutes Cycle Travel time penalty for Cycle > 4000 metres 30 minutes

Appendix F

Public Transport fare rates

Table F-1: TransPerth public transport fare rate

Number of TransPerth Zones Fare ($) 1 2.5 2 3.7 3 4.6 4 5.4 5 6.6 6 7.6 7 8.7 8 9.5 9 10.2 (Source: TransPerth 01 July 2010, www.transperth.wa.gov.au)

Table F-2: Public transport fare coding for TODTrips

Origin Destination Fare ($) within study area within study area 2.5 within study area NEDLANDS 2.5 within study area CLAREMONT 2.5 within study area WANNEROO 6.6 within study area MUNDARING 5.4 within study area SWAN 4.6 within study area KALAMUNDA 4.6 within study area SERPENTINE-JARRAHDALE 6.6 within study area ARMADALE 5.4 within study area ROCKINGHAM 6.6 within study area KWINANA 5.4 within study area COCKBURN 4.6 within study area MELVILLE 3.7 within study area FREMANTLE 3.7 within study area EAST FREMANTLE 3.7 within study area CANNING 3.7 within study area SOUTH PERTH 2.5 within study area VICTORIA PARK 2.5 within study area BELMONT 2.5 within study area VINCENT 3.7 within study area BAYSWATER 3.7 within study area PERTH 2.5 within study area KINGS PARK 2.5 within study area SUBIACO 2.5 within study area STIRLING REMAINDER EAST 2.5 within study area CAMBRIDGE 3.7 within study area MOSMAN PARK 3.7 within study area PEPPERMINT GROVE 3.7 within study area COTTESLOE 3.7 within study area JOONDALUP 4.6 within study area GOSNELLS 4.6 within study area BASSENDEAN 3.7 within study area STIRLING REMAINDER WEST 2.5 within study area MANDURAH 8.7 within study area MURRAY 8.7