Cooperative Network Design

EVAIR Safety Bulletin No 4

Summer Seasons 2006-2009 Evolution

EUROCONTROL

EVAIR_bulletin04.indd 1 15/02/10 14:15 EUROCONTROL VOLUNTARY ATM INCIDENT Data providers REPORTING (EVAIR) The number of EVAIR ATM incident data providers among commer- cial airlines has been increasing year on year. By the end of 2009, 63 SUMMER SEASONS 2006-2009 EVOLUTION airlines were providing ATM incident reports to EVAIR. The airlines which contribute to EVAIR account for 50% of the overall European EXECUTIVE SUMMARY air traffic.

The statistics in the EVAIR Safety Bulletin No 4 cover the summer Main trends periods (April - September) 2006 to 2009. The statistics are based Eight main ATM contributors are used as the basis for monitoring: on the ATM/CNS incident reports collected manually and ACAS RAs collected automatically from one Mode-S radar station. The 1. Mistakes; main data providers are commercial airlines and ANSPs providing 2. Spoken Communication; feedback to airlines’ incident reports and the ACAS RAs from the 3. Operational Communication; Mode-S radar station. 4. Traffic and Airspace problems; 5. Traffic Information; Data collection 6. Lapses; Through the manual data provision (commercial airlines and feed- 7. ATC Clearance; back from ANSPs) EVAIR collected approximately 1500 reports for 8. Coordination. the summer seasons from 2006 to 2009. For the same period, 2637 valid ACAS RAs were collected automatically from the Mode S radar During the summer periods 2006-2009 in absolute figures, Mistakes station. with 27%, Spoken communication with 19%, Operational Communi- cation and Traffic information with 11% each make almost 70% of the The number of reports EVAIR collects has grown each year. In per- overall ATM contributors. centage terms the increase is the following: 2007 – 26% more than in 2006; 2008 - 28% more than in 2007; and 2009 - 78% more than EVAIR Safety Bulletin - Airlines’ Contribution in 2008. In this EVAIR Safety Bulletin we introduce a new ‘Airlines’ Corner’ section and Brussels Airlines is our first contributor. They are one It is emphasised that there is no evidence to suggest that the of the most pro-active EVAIR data providers, and we thank increase in the number of the reported incidents is reflective of them for their kind offer to share their safety data with the avia- an erosion of the ATM safety performance but rather it is due to tion safety community. the increase of voluntary reporting as promoted by EVAIR, familia- risation of the ATM incident data providers with these activities Security and Confidentiality and a general improvement of the reporting culture. In collecting and processing data, EVAIR follows strict security and confidentiality arrangements. Safety data provided are properly safe- Among six phases of flight (landing, standing, taxiing, take-off, guarded and de-identified and the information is only used for the approach and en-route) for the summer periods 2006-2009, the promotion and enhancement of aviation safety. statistics (in absolute figures) show that the largest number of inci- dents (78%) occur within the en-route and approach phases. In the EVAIR Suggestions/Improvements summer periods 2006 - 2008, the largest number of incidents was EVAIR is constantly looking at ways to improve its services and within en-route phase; however, during summer 2009, the biggest products and suggestions and proposals are more than welcome. number of incidents occurred within the approach phase. Indeed, Please forward any thoughts, ideas and comments to Ms Dragica during summer 2009 there were 1.5 incidents per 10,000 flight op- Stankovic, EVAIR Function Manager: [email protected] erations in the approach phase compared with 1.1 incidents during the same period in 2008. © Skyguide

EUROCONTROL 2 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 2 15/02/10 14:15 CONTENTS

Executive Summary 2

Contents 3

EVAIR Function Manager’s Perspective 4

Introduction to EVAIR Statistical Data 6

Section 1 - Phases of Flights 7

l Main contributors to ATM incidents 8

Section 2- EVAIR Support to European Action Plans 13

l Runway Incursions 14 l Level Busts 15

Section 3 - EVAIR Support to Call Sign Similarity Implementation Project 17

Section 4 - ACAS Monitoring 20

l Part One: Manual 20 l Part Two: ACAS RAs Collected Automatically From One Mode-S Radar Station 25

Section 5 - Miscellaneous 34

l Lasers 34

EVAIR Airlines’ Corner 35

Annexes 42

l Annex 1 - European Action Plans 42 l Annex 2 - Definitions & Acronyms (from Heidi and HERA Taxonomies) 43 l Annex 3 - Acronyms 46

EVAIR Team 47

EVAIR Summer Seasons 2006-2009 Evolution 3 EUROCONTROL

EVAIR_bulletin04.indd 3 15/02/10 14:15 EVAIR FUNCTION MANAGER’S

PERSPECTIVE

We are very pleased to say that thanks to our stakeholders’ agreed that the timely provision of the feedback improves the voluntary contributions and their commitment to EVAIR activi- communication and more importantly develops trust between ties, we have had consistent year on year growth in the number airlines and ANSPs. The role of EVAIR is to facilitate the feedback of collected incident reports. The graph below shows these process by keeping up to date safety focal points on the side of the increases of manually reported incidents in absolute figures. airlines and ANSPs; forwarding ATM incidents to ANSPs or airlines involved; asking for the reply and, when/if the feedback arrives, forwarding it to an appropriate address. The majority of the 700 600 requests for the feedback are still initiated by the airlines but the 500 communications channels are open two-ways and EVAIR would 400 be happy to act as a conduit from ANSPs to airlines if requested 300 to do so. 200 100 Spread of incidents through phases of flights 0 2006 2007 2008 2009 The graph below shows the spread of ATM incidents through phases of flight, (i.e. standing, taxiing, take off, approach and Figure 1 – Incident data collection (absolute figures) en-route). Through summer periods 2006 - 2009, the absolute figures among six phases of flight, show that the largest number Data providers of incidents (78%) occurred within En-route and Approach phases. It is more than clear that without data providers there is no EVAIR. Until 2008, the largest number of incidents occurred within We are very glad to say that our main stakeholders - commer- En-Route phase. However, data for summer 2009 shows that the cial airlines and ANSPs - recognise EVAIR as an activity which is largest number of incidents occurred within the Approach phase, addressing their interests. EVAIR continues its efforts to enlarge which is close to the trends identified by the other safety data bases. the team of providers among Airspace Users and Air Navigation Service Providers. Indeed during summer 2009 there were 1.5 incidents per 10, 000 flight operations in the Approach phase compared with 1.1 incidents Automated data collection during 2008. More detailed statistics follow. In the future we expect more incident data to be collected auto- Standing matically. In this regard, EVAIR is making all necessary prepara- Landing 0.9% Take-off 0.7% tions to be able to process and analyse this larger amount of data. 10.3% We are working on the enlargement of the number of Mode-S En-route radars from where ACAS RAs will be collected and also on the 37.8% Taxiing 10.5% training developments for the ANSPs to use new tools for auto- 700 mated600 data collection and analysis. 500 Feedback400 – Motivation for better reporting As300 one of the main pillars of EVAIR activities, the EVAIR feedback mechanism200 is becoming increasingly recognised by all stake- 100 holders as a means to help airlines (and ANSPs) address issues and 0 2006 2007 2008 2009 provide quick fixes. This also has important benefits in that it acts Approach as an incentive to pilots to report more. During recent meetings 39.8% between EVAIR and some of the ANSPs and the airlines, it was Figure 2 – Incident per phases of flight Summers 2006 – 2009

EUROCONTROL 4 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 4 15/02/10 14:15 Eight main ATM contributors have been used as the basis for within 10 different European states. More anecdotal evidence and monitoring: press/media stories confirm this unwelcome trend. The problem cannot be resolved by the aviation alone rather it requires 1. Mistakes concerted interventions involving amongst others the police, 2. Spoken Communication judiciary, manufactures and law makers. For its part, EVAIR may 3. Operational Communication investigate the feasibility of promoting wider European actions to 4. Traffic and airspace problems try and stop the escalation of this problem. 5. Traffic information 700 6. Lapses New section – Airlines’ corner 600 7. ATC Clearances Within this EVAIR Safety Bulletin issue we introduce a new section 500 8. Coordination dedicated to our incident data providers. This innovation provides 400 300 an opportunity for our stakeholders to present their Safety 200 During the summer periods 2006-2009, in absolute figures, four Management and data collection systems and share with us their 100 contributors out of eight (Mistakes, Spoken communication, concerns and findings which can be compared with those identi- 0 2006 2007 2008 2009 Operational communication and Traffic information) account for fied by EVAIR. almost 70% of the overall ATM contributors.

Traffic and Airspace We are delighted that Brussels Airlines has agreed to make the problems ATC clearance/ Operational first contribution. It is as one of the most pro-active EVAIR data 6% Instructions comms 10% 11% providers and we thank them for their kind offer to share their

Coordinations Spoken comms safety data with the aviation safety community. issues 19% 7%

Traffic information 11%

Lapses Mistakes 9% 27%

Figure 3 – Contributors to ATM incidents Summer 2006 – 2009

Customised statistics – Laser, Call Sign Confusion Besides the contributors listed above, EVAIR also produces statis- tics for some specific items such as Call Sign Confusion and Laser interference in aviation. The latter issue was first featured in EVAIR Safety Bulletin No 3. Since then we have seen a further increase in the number of reports related to laser interference and it is clear that this is a growing problem and concern across aviation. EVAIR data shows that laser issues have been identified at 16 locations

EVAIR Summer Seasons 2006-2009 Evolution 5 EUROCONTROL

EVAIR_bulletin04.indd 5 15/02/10 14:15 INTRODUCTION TO STATISTICAL DATA

Statistics as presented in EVAIR Safety Bulletins are based on inci- EVAIR Safety Bulletins are issued twice per year. One issue (like dents received from commercial aircraft operators and feedback this one) covers the summer season (April - September) whilst the data provided on some of them by ANSPs. Incident reports are other covers the whole year period. Data are provided by more very much based on the subjectivity of those who were involved than 60 airlines. When solicited, all ANSPs have provided feedback and in the first place pilots who filed the report and described the to airline reports. occurrence. EVAIR statistics do not contain severity analysis, since the analysed reports are not officially investigated or the Note: In the EVAIR Safety Bulletin within the manual reporting official (airline/state) investigation is still waiting to be closed. section we give either absolute figures of reported occurrences Nonetheless, the statistics do provide a general view and show or the number of reported occurrences per 10,000 flight opera- some main trends of the current operational safety acceptability. tions. Some pie charts show two figures: one figure shows the percentage and the other one rate of incidents per 10,000 flights. EVAIR activity covers the whole ECAC airspace as well as some of Within the automated TCAS RA data collection section all figures the airspaces neighbouring the ECAC region. In the same way, the are in absolute values. airlines and ANSPs who participate in EVAIR come from across Europe and, indeed, some airlines based outside of Europe but Definitions: Definitions for each element contained in the graphs who regularly fly through European airspace. can be found in the Annex 2.

EUROCONTROL 6 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 6 15/02/10 14:15 SECTION 1 - PHASES OF FLIGHTS

INCIDENTS PER PHASE OF FLIGHT: 2006 - 2009

Figure 4: Incidents per phases of flight. Approach Summer 2006 - 2009 (absolute figures) At all phases of flights during summer 2009 there was En-route a significant increase in the number of incidents. It is believed that this increase of reported incidents is not Landing reflective of a degradation of the ATM safety performance but is most likely due to the constant enlargement of the 2009 2009 Standing EVAIR incident data providers. 2008 2008 2007 Relations between the different phases of flights Take-off 2006 through the period of the four summer seasons show that the Approach and En-route phase have a consis- Taxiing tently higher number of incidents. In 2009 in the

0 50 100 150 200 250 300 Approach phase there0.0 were0.2 many0.4 more0.6 incidents0.8 than1.0 1.2 1.4 1.6 in En-route, which is a reversal of the previous 3 summer seasons when there was a slightly higher number of incidents in the En route phase.

2009 Figure 5: Incident distribution per phases of flight. Approach 2008 Summer 2008 - 2009 per 10,000 flight operations 2007 As anticipated from Figure 2, the En-route and Approach 2009 En-route 2006 phases contributed for more than 3/4 of the total number 2008 of reported occurrences. Landing During the summer season 2009 the Approach phase 0 50 100 150 200 250 300 had higher rate than0.0 En-route0.2 with0.4 approximately0.6 0.8 1.5 1.0 1.2 1.4 2009 Standing 2009 2008 2008 incidents per 10,000 flights. 2007 Due to the short period of data collection, it is still prema- Take-off 2006 ture to assert exactly how close this distribution is to reality.

Taxiing

0 50 100 150 200 250 300 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

2009 EVAIR Summer Seasons 2006-2009 Evolution 7 EUROCONTROL 2008 2007 2009 2006 2008

EVAIR_bulletin04.indd 7 15/02/10 14:15

0 50 100 150 200 250 300 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 MAIN CONTRIBUTORS TO ATM OCCURRENCES

The charts that follow show the list of eight ATM occurrence contributors’ statistics for summer seasons 2006-2009 in absolute figures and summers 2008-2009 comparison per 10,000 flights. The eight contributors were identified as those which most frequently caused different 2009 2009 types of occurrences. The graphs show the distribution of the main contributors for all phases. Furthermore, there are separate2008 graphs 2008 for the En-Route and Approach phases due to high proportion of occurrences within those two phases. The intention is to highlight2007 any 2006 trends that may be apparent between the different contributors and the flight phases. The definitions of all contributors can be found in Annex 2.

The eight Main ATM Contributors for all phases of flight. 0 50 100 150 200 250 300 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

Figure 6: Contributors to ATM incidents. Operational comms Summer 2006 -2009 (absolute figures) Spoken comms During summer 2009, Mistakes and Spoken Communication were the most common Mistakes contributors for all phases of flight. More- Lapses over, they had much higher increases than other contributors which could be explained Traffic information 2009 2009 2009 by the increase of incident data providers 2008 2008 Coordination 20072008 and better reporting. Coordination issues 2009 2007 2006 2008 also had significant increase in 20092006 while ATC clearance/Instructions Lapses and Traffic information continued to Traffic and Airspace problems decline. 0 50 100 150 200 250 300 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0 50 100 150 200 250 300 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

Figure 7: Contributors to ATM incidents to Operational comms all phases of flights - Summer 2008-2009 per 10,000 flight operations Spoken comms During the summer 2009, three contribu- Mistakes tors showed very high increase vs 2008: Mistakes, Spoken communication and Lapses Coordination. In all other areas, in spite of 2009 Traffic information the increase of data providers, the trend2008 is 2007 Coordination decreasing. Mistakes contributed to almost 2009 2006 1.4 incidents per 10,000 flights during 2008 ATC clearance/Instructions summer 2009, which is more than twice as many as summer 2008. Traffic and Airspace problems 0 50 100 150 200 250 300 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

More detailed statistics for contributors’ distribution against the En-route and Approach phases of flight follow. A selection of the 3 main contributors, i.e. Mistakes, Spoken Communications and Coordination issues, will be further broken down to assist the reader’s understanding.

EUROCONTROL 8 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 8 15/02/10 14:15 Distribution of Main Contributors for En-route and Approach Incidents

Figure 8: Contributors to ATM incidents Operational comms for En-route phase. Spoken comms Summer 2006 - 2009 (absolute figures) Within the En-route phase, during summer Mistakes 2009 as for all phases of flight, the same 3 Lapses contributors, Mistakes, Spoken communi- cation and Coordination issues showed an Traffic information 2009 2009 20082009 20082009 upward trend. Coordination 20072008 20072008 EVAIR will continue monitoring these three 20062007 20062007 areas in the coming years. ATC clearance/Instructions 2006 2006

Traffic and Airspace problems

0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80 90 100

Figure 9: Contributors to ATM incidents in Operational comms Approach phase. 2009 Spoken comms Summer 2008 - 2009 (absolute figures) 20082009 In the Approach phase, the same three 2008 Mistakes contributors (Mistakes, Spoken communi- cation and Coordination) issues, showed Lapses 2009 an even greater increase than in the Traffic information 20082009 En Route phase. Coordination issues, 20072008 showed a 7-fold increase in summer 2009 Coordination 20062007 2006 compared with the same period in 2008. ATC clearance/Instructions In addition, Traffic and Airspace also Traffic and Airspace problems increased in prominence in the Approach 0 20 40 60 80 100 120 140 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 phase. 0 20 40 60 80 100 120 140 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50

EVAIR Summer Seasons 2006-2009 Evolution 9 EUROCONTROL

EVAIR_bulletin04.indd 9 15/02/10 14:15 Drill down through “Spoken Communications” contributor (see definition in Annex2)

Figure 10: Spoken Communication. Call sign confusion Summer 2006 – 2009 (absolute figures)

Out of eight measured areas (see left hand of High R/T workload graph) five show an increase (2 significantly), language/accent while three decreased during summer

2009. ‘Situation not conveyed by pilots’ had Misunderstanding/interpretation the largest number of incidents. From the 2009 Noise interference 2009 increase perspective, ‘Situation not conveyed 2008 2008 by pilots’ and ‘Poor/no coordination’2007 had Pilots breach of R/T 2007 almost the same increase, i.e. they show20062009 six 20062009 2008 Poor /no coordination 2008 times more incidents during summer 2009 2007 2007 compared with the same period in 2006 2008. Situation not conveyed by pilots 2006 0 10Call sign20 confusion30 40was another50 area60 which70 80 0 10 20 30 40 50 60 70 80 90 100 increased.

0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80 90 100

Figure11: Spoken Communication. 2009 Summer 2008 - 2009 per 10,000 flight Call sign confusion 2008 operations. High R/T workload 2009 In the Approach phase, the same three 2008 language/accent contributors (Mistakes, Spoken commu2009 - nication and Coordination) showed2008 an Misunderstanding/interpretation even greater increase than in the En 2007Route 20062009 phase. Coordination issues showed a 7-fold Noise interference 2008 increase in summer 2009 compared2007 with Pilots breach of R/T the same period in 2008. In addition, 2006Traffic 0 20and Airspace40 also60 increased80 in100 prominence120 140 Poor /no coordination0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50

in the Approach phase. Situation not conveyed by pilots 0 20 40 60 80 100 120 140 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50

EUROCONTROL 10 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 10 15/02/10 14:15 Drill down through ‘Mistakes’ contributor (see definition in Annex2)

Figure 12: Drill down through Mistakes Decision making for Approach/En-route of flights. Summer 2006-2009 Failure to monitor During summers 2006 - 2009, Judgment Information wrongly associated

and Planning were the highest contributors Judgement to the Mistakes which occurred within the Mental/Emotional/ Personnality issues Approach and En-route phases. There were Mis stored or Approach Approach insufficiently learned information nearly twice as many Planning contributions En-route En-route to the Mistakes in the Approach phase than Planning in the En Route phase. Recall of information 0 20 40 60 80 100 120 140 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

Violations Approach Approach En-route En-route Workload issues 0 20 40 60 80 100 120 140 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

Figure 13: Drill down through Mistakes for Approach/En-route. Summer 2008 per Decision making 2009 10,000 flight operations Approach En-route 2008 In 2008 for Approach/En-Route, four Information wrongly associated 2007 contributors to Mistakes were identified 2006 Judgement of which Planning and Judgment had Approach 2009 the highest rate; Planning being a higher Planning En-route 2008 problem within En-Route and Judgment 2007 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0 5 10 15 20 200625 30 within the Approach phase.

0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0 5 10 15 20 25 30

Figure 13a: Drill down through Mistakes Decision making for Approach/En-route. Summer 2009 per 10,000 flight operations Failure to monitor In 2009, eight contributors to Mistakes Information wrongly associated were identified. Judgment was the highest one and, as in 2008, it was much Judgement

higher within Approach than within the Mental/Emotional/ Personnality issues En-Route phase. Planning

Recall of information Approach Approach En-route Workload issues En-route

0 20 40 60 80 100 120 140 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

EVAIR Summer Seasons 2006-2009 Evolution 11 EUROCONTROL

EVAIR_bulletin04.indd 11 15/02/10 14:15

Approach 2009 En-route 2008 2007 2006

0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0 5 10 15 20 25 30 After four years of data collection and analysis, EVAIR has made several data and trend validations with a number of other stakeholders (ANSPs and Airlines) contributing to EVAIR activities. It is very encouraging to see that compared areas have the same or very similar trends. For further validation of the EVAIR statistics and identified trends, Airlines and ANSPs are kindly invited to compare them with their own statistics and let EVAIR know whether these trends are in line, or not, with ours.

EUROCONTROL 12 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 12 15/02/10 14:15 SECTION 2 - EVAIR SUPPORT TO EUROPEAN ACTION PLANS

One of EVAIR’s objectives is to support European Safety Action Plans by monitoring the main contributors to safety occurrences for Level Bust, Runway Incursions & Air –Ground Communications as identified by those Action Plans (see Annex 1). The monitoring aims to identify trends of increase or decrease of the main causal factors as defined by the Safety Action Plans and to warn if new causals appear. In general it should assist in prioritising actions which should be taken on a pan-European level.

The following graphs show statistics for ATC clearance instructions where, among others, are Runway Incursion and Level Bust (see definition in Annex2). Again dataApproach relates to the summer periods 2006 to 2009. Approach En-route En-route

0 20 40 60 80 100 120 140 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

Figure 14: ATC Clearance Instructions. Summer 2006 – 2009 (absolute figures) Level Bust

In spite of the large general increase of Assigned route/track/heading

collected incident reports, in the summer Assigned Speed 2009 all events related to ‘ATC Clearance Climb/descent conditional clearance instructions’ show a significant decrease. The number of Level Bust incident reports in En-route clearance

the summer 2009 was two times lessApproach than in Holding 2009 2008 the same period of 2007 and 2008.En-route Runway Runway incursion 2007 Incursion had the second largest drop with Taxiway incursion 2006 many fewer incidents than in summer 2007. Rate of climb/descent restrictions A new event identified in summer 2009 is Missed Approach. Missed approach 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0 5 10 15 20 25 30

Figure 15: ATC Clearance Instructions. Summer 2008 - 2009 per 10,000 flight Level Bust

operations Assigned route/track/heading In line with Fig 14 data above, all events Assigned Speed related to the ATC Clearance instructions show a decrease vs summer 2008 when Climb/descent conditional clearance

measured per 10,000 flights. During En-route clearance summer 2009 there were 0.07 Level Busts 2009 Rate of climb/descent restrictions 2008 per 10,00 flights and 0.03 Runway Incur- 2007 Missed approach sions incidents. Missed Approaches are 2009 2006 events which appeared in EVAIR summer Runway incursion 2008 2009 statistics for the first time. Their level Taxiway incursion is close to that of Taxiway and Runway 0.00 0.05 0.10 0.15 0.20 0 2 4 6 8 10 12 Incursion incidents.

EVAIR Summer Seasons 2006-2009 Evolution 13 EUROCONTROL

2009 EVAIR_bulletin04.indd 13 15/02/10 14:15 2008 2009 2008

0.000 0.005 0.010 0.015 0.020 0.025 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 RUNWAY INCURSIONS

EVAIR data for the summer periods 2006 - 2009 show a decreasing trend in the number of Runway Incursions. Summer 2007 had the highest number of Runway Incursions while summers 2008 and 2009 show a significant decrease. It is worth noting that the decreasing number of recorded Summer 2009 Runway Incursion incidents is against the background of a significant rise in the number of reports: in summer 2009 EVAIR collected an additional 70% of incident reports compared with summer 2008 and this figure rises to 80% for summer 2007. Therefore the summer 2009 figures look encouraging, but it is necessary to highlight that we would like to verify EVAIR data with other data bases to be able to confirm these positive trends. We can report that, in the latest SRC Annual report, comparison between 2007 and 2008 shows a reducing trend of Runway Incursions similar to that identified by EVAIR data.

Figure 16: Runway Incursions. Oper. Comm. Summer 2006 - 2009 main contributors (absolute figures) Spoken comm. Two contributors, Mistakes and Spoken communication had a slight increase Mistakes compared with the summer 2008. It is 2009 2008 encouraging that they are much lower than 2009 Lapses 2007 summer 2006. 2008 2009 2006 2007 2008 2009 Traffic information 2006 2008

Coordination 0.00 0.05 0.10 0.15 0.20 0 2 4 6 8 10 12 0.00 0.05 0.10 0.15 0.20 0 2 4 6 8 10 12

Figure 17: Runway Incursions main con-

tributors. Summer 2008 – 2009 per 1,000 Oper. Comm. flight operations Spoken comm. In summer 2009, out of five contributory 2009 areas, two of them, Spoken communi- Mistakes 2008 20092009 cation and Mistakes, show the highest 20082008 2009 Lapses number of incidents while all other areas 2008 show a significant decrease compared Traffic information

with 2008. 0.000 0.005 0.010 0.015 0.020 0.025 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.000 0.005 0.010 0.015 0.020 0.025 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14

EUROCONTROL 14 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 14 15/02/10 14:15 PREVENTION OF LEVEL BUSTS 2009 2008 2007 EVAIR data for Level Busts in summer 2009 compared with the same periods in 2007 and 2008 shows a significant decrease.2006 As stated previously this is encouraging when viewed2008 against a rising number of received reports, but again EVAIR would like to verify these data with other European data bases to be able to confirm these positive trends.

0.00 0.05 0.10 0.15 0.20 0 2 4 6 8 10 12

Figure 18: Level Bust distribution per phases of flight. Summer 2008-2009 per 10,000 flight operations Summer 2009 data shows a significant

decrease in the number of Level Busts Approach 2009 2008 compared with the same period in 2008.2009 During summer 2009, the number of Level2008 En-route Busts was the same within the Approach Take-off phase while during summer 2008 a larger 0.000 number0.005 of Level0.010 Busts occurred0.015 within0.020 the 0.025 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 En-route phase.

Oper. Comm.

Figure 19: Main contributors to Level Spoken comm. Busts. Summer 2006-2009 (absolute figures) Mistakes All Level Bust contributors in summer 2009 2009 show a significant decrease 2008 Lapses compared with the previous three years. If looking only at summer 2009, Spoken 2009 Traffic information 2008 communication and Mistakes are the 2007 most prominent contributors compared Coordination 2006 with the other contributors. 0 2 4 6 8 10 12 0.00 0.02 0.04 0.06 0.08 0.10

EVAIR Summer Seasons 2006-2009 Evolution 15 EUROCONTROL 2009 2008 2007 2006 EVAIR_bulletin04.indd 15 15/02/10 14:15 0 1 2 3 4 5 6 Figure 20: Level Bust contributors. Oper. Comm. 2009 Summer 2008-2009 per 10,000 flight 2008 operations Spoken comm. In summer 2009, all Level Bust contribu- 2009 Mistakes tors identified by EVAIR show a significant2008 decrease. Spoken communication with2007 0.02 Lapses 2006 incidents had the highest level. Traffic information 2009 0 2 4 6 8 10 12 0.00 0.02 0.04 0.06 0.08 0.10 2008

2009 2008 2007 2006

0 2 4 6 8 10 12 0.00 0.02 0.04 0.06 0.08 0.10

Figure 21: Drill down through Spoken Call sign confusion communication contributors to Level2009 High R/T workload Bust. 2008 2007 Summer 2006 - 2009 (absolute figures) 2006 language/accent Total The number of the reported Level Busts 0 during1 summer2 20093 is significantly4 5 lower 6Misunderstanding/interpretation than for the previous period. The impres- Noise interference sion is that attention should still be paid to 2009 Misunderstanding/interpretation contribu- Poor /no coordination Total 2008 tors and Language/accent since they also 2007 2006 contribute to other types of incidents. Situation not conveyed by pilots

0 1 2 3 4 5 6

EUROCONTROL 16 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 16 15/02/10 14:15 SECTION 3 - EVAIR SUPPORT TO CALL SIGN SIMILARITY IMPLEMENTATION PROJECT

General A EUROCONTROL Call Sign Similarity (CSS) project was initiated in April 2008 with the aim of establishing pan-European CSS solutions centred on a coordinated service operated by EUROCONTROL’s CFMU.

A stepped-approach is envisaged:

l Service Level 0 involving the establishment of a Call Sign Similarity Management Cell (CSMC) within EUROCONTROL’s CFMU to raise awareness and the publication of call sign similarity rules. l Service Level 1 providing Aircraft Operators (AOs) with a CSS Service and associated CSS Tool to detect and then de-conflict call signs within their individual schedules (Winter 2011). l Service Level 2 will provide de-confliction between multiple AO schedules.

Monitoring Principles During the implementation and post-implementation period, the EVAIR function will be used to monitor the tactical safety performance aspects of the CSS Service/Tool. The basic principles and reporting lines have been agreed within the Call Sign Similarity User Group (CSS UG), the CFMU Call Sign Similarity Management Cell (CSMC) and the EVAIR Team (see schematic below).

ANSPs EUROCONTROL Aircraft operator (using CSS tool)

ATCO CSMC Pilot Liaison only Similatity (resolution and Confusion of CSS) ‘Analysis’ Ops/Safety Similatity Ops/Flight Scheduling and Confusion Noti
cation Reports Internal & Liaison Feedback Similarity & Liaison / Action (on request) Confusion EVAIR Safety Reports & Feedback AST Safety (Confusion) ‘Analysis’ Confusion Reports (via EVAIR)

Confusion Reports & Feedback Aircfraft operator (not using CSS tool)

Figure 22: Call Sign Similarity Tool Service Level 1 - Tactical Call Sign Similarity and Safety (Confusion) Performance Monitoring Scheme

Essentially, AOs (using the CSS Tool) and ANSPs will be encouraged to report tactical call sign similarity and call sign confusion events to EVAIR who will forward them internally to the CSMC. The CSMC will then analyse the reports using the metrics (described later) and will provide any feedback to EVAIR who in turn will send this to the AOs and ANSPs via the existing EVAIR channels. The CSMC may also converse directly with AOs when it is necessary to do so for instance when there is a need for AOs to resolve a particular call sign issue. ANSPs can also contact AOs directly (in particular those who are not using the CSS Tool) if they so wish. The exact mechanics, i.e. the com- munications means to be used (email, fax etc), frequency of reporting, format of reports etc is being considered and will be made known well ahead of the CSS tool implementation phase.

EVAIR Summer Seasons 2006-2009 Evolution 17 EUROCONTROL

EVAIR_bulletin04.indd 17 15/02/10 14:15 Monitoring Aims The monitoring will help to:

l Trace the impact of safety incidents and any remedial the measures taken. l Demonstrate an improving safety situation (fewer reported ‘call sign confusion incidents’). l Verify the performance and effectiveness of the CSS Tool/Service.

Monitoring Metrics To support the monitoring, the following Metrics may be used:

l Comparison of the number of similar call signs within a single AO’s schedule before detection and de-confliction by the CSS Tool and the number of similar call signs after use of the CSS Tool. This is part of the ‘strategic’ monitoring and is based around the set of Call Sign Similarity ‘Rules’ established by the Call Sign Similarity User Group. l The number of tactical safety events reported to, and processed by, EVAIR/CSMC. l The number of tactical call sign confusion occurrences reported versus the number of flights by an AO within European (ECAC) airspace. l The number of tactical call sign confusion reported occurrences per 10,000 flights. l Comparisons of actual tactical results against the anticipated reduction of Call Sign Confusion occurrences.

Monitoring - CSS Tool Trial It will be particularly important to enlist the help of ANSPs to monitor the tactical call sign similarities and confusions during the trial period of the CSS Tool which is currently envisaged to take place from late 2010 until mid 2011. It is anticipated that a small number of AOs will enrol on the trial programme. Assuming that the CSS Tool enables them to resolve most if not all of the similar call signs embedded in their schedules, we will need ANSPs to inform us of actual similarity or confusion events involving 2 or more aircraft within the same company for the airlines taking part in the trial. This information will be used to fine-tune the parameters, configurations and settings built into the CSS Tool before the full operational roll-out in late 2011.

Call Sign Similarity Rules In the meantime, the CSMC aims to publish the agreed set of similarity rules on the CSMC web site. This will enable AOs who are not already using some form of call sign detection/de-confliction activity to have a basis upon which they can start without having to wait for the implementation of the CSS Tool.

CSS Project Information More information concerning the CSS Project (e.g. how the CSS Tool development is progressing) can be found on the EUROCONTROL CND website pages at http://www.eurocontrol.int/safety/public/standard_page/Callsign_Similarity_project.html . Operational CSS information, including the CSS Rules, will be available via the CSMC web pages and CFMU NOP.

CSMC - DSNA Monitoring Since September 2009, the CSMC has been responsible for running the ATC call sign similarity detection and de-confliction service previously undertaken by France’s DSNA on behalf of a small number of (mainly French) airlines. This work is provid- ing the CSMC staff with valuable insights into CSS operations, problems etc and lessons learnt are being fed into the development of the EUROCONTROL CSS Tool.

Moreover, the CSMC (and EVAIR) are now receiving call sign similarity/confusion reports from the AOs participating in the DSNA scheme. Whilst there are still problems, the detection and de-confliction rate is impressive and other airlines are keen to join the scheme. However, the DSNA call sign similarity tool is very labour intensive and there is no scope for further expansion. This provides additional impetus for development and deployment of a user-friendly CSS Tool that will be accessible (from the CFMU NOP Portal) for all aircraft operators who wish to take part.

EUROCONTROL 18 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 18 15/02/10 14:15 Call Sign Confusion versus phases of flights

Figure 23: Call Sign Confusion per phases 14 16

of flight. Approach 12 14 Mistakes Summer 2006-2008 (absolute figures) En-route Spoken Communications In general summer 2009 shows a higher 12 Oper. Communications 10 number of call sign confusions. The highest ATC clearance / Instructions 10 number has occurred within the En-Route 8 phase. 8 6 6 4 4

2 2

0 0 2006 2007 2008 2009 2006 2007 2008 2009

14 Figure 24: Call Sign Confusion ATM 16

Contributors.Approach Summer 2006 – 2009 12 14 Mistakes (absoluteEn-route figures) Spoken Communications 12 Oper. Communications 10 Spoken communication through the moni- tored period has consistently increased ATC clearance / Instructions 10 2009 8 with a further escalation during summer 2008 2009 2009. The significant increase in summer 8 6 2008 2009 could be explained by an overall 6 2007 increase of incident data collection and 2006 4 ongoing CSS project implementation and 4 0 20 40 60 80 100 120 140 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 2 wider awareness of the problem. 2

0 0 2006 2007 2008 2009 2006 2007 2008 2009

2009 2008 2009 2008 2007 2006 EVAIR Summer Seasons 2006-2009 Evolution 19 EUROCONTROL 0 20 40 60 80 100 120 140 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

EVAIR_bulletin04.indd 19 15/02/10 14:15 SECTION 4 - ACAS REPORTING

EVAIR ACAS monitoring aims at the identification of causals and their trends in order to take proper preventive measures to ensure the continued safe operation of ACAS.

ACAS is the generic term for Airborne Collision Avoidance Systems, of which TCAS II is the only example so far. ACAS is intended to improve air safety by acting as a ‘last-resort’ method of preventing mid-air collisions or near collisions between aircraft. Although the implementation of TCAS phase 2 has been terminated, ACAS monitoring continues to improve safety by identifying technical and procedural deficiencies.

ACAS data have been collected either automatically via the Automated Safety Monitoring Tool (ASMT), developed by EUROCONTROL, or manually thanks to airlines and ANSPs reporting. Statistics given in Part One of this section are derived from manual reporting. Data collected automatically from only one Mode S radar station provided by one of14 the volunteering ANSPs are provided in Part Two. 16

Approach 12 14 Mistakes For summer periods 2006-2009, EVAIR collected, through manual reporting, about 350En-route TCAS RA reports for the whole ECAC airspace; Spoken Communications 12 Oper. Communications however, through the automatic data collection from only one Mode-S radar10 station for the same period we received 2637 Valid RA Messages. This comparison is one of the best indicators of what is actually happening and underlines the importance of automatic ATC clearance / Instructions 10 reporting. It also sheds light on the current level of manual ATM incident reporting8 and where we should focus to have more data. 8 6 In addition, it should be noted that ACAS / TCAS statistics from manual reporting are the pilots’ perception of the events rather 6 than measured or calculated. Therefore care is needed when comparing manually collected data and data that is captured auto- 4 matically and messages about performance should generally be taken from the Automatic recording of all events. 4

2 2

0 0 2006 2007 2008 2009 2006 2007 2008 2009 PART ONE - MANUAL ACAS REPORTING

Figure 25: Manually reported ACAS incidents by phase of flight. Summer 2006 - 2009 (absolute figures) The summer periods 2006-2009 show a consistent increase in the number of manu- Approach 2009 ally reported ACAS incidents. The En-route 2008 2009 phase is the most affected. Moreover, the En-route 2008 number of ACAS incidents has a higher rate 2007 than other types of incidents. Given the Take-off 2006 high prominence of ACAS incidents, they deserve serious attention in order to iden- 0 20 40 60 80 100 120 140 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 tify as far as possible more causals so that improved mitigations may be put in place to reduce the incidence of them.

EUROCONTROL 20 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 20 15/02/10 14:15 14 16

Approach 12 14 Mistakes En-route Spoken Communications 12 Oper. Communications 10 ATC clearance / Instructions 10 8 8 6 6 4 4

2 2

0 0 2006 2007 2008 2009 2006 2007 2008 2009

Figure 26: Manually reported ACAS

incidents by phase of flights. Summer Approach 2009 2008 - 2009 per 10,000 flight operations 2008 2009 Within manual ACAS reporting, the largest 2008 En-route number of ACAS incidents occurs during2007 the En-Route phase. In summer 2009 this2006 Take-off rate vs 10, 000 operations was slightly lower 0 20 40 60 80 100 120 140 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 than in summer 2008.

Figure 27: ACAS RA Classification. Summer 2006 -2009 (absolute figures) Unclassifiable RA 2009 All summer periods show that the largest 2009 2008 2008 number of ACAS occurrences has been 2007 Nuisance classified by pilots as ‘Useful RAs’. Even 2006 those classified as ‘Nuisance’ does not mean Useful RA that they are not ‘useful’. They are clearly

defined and the definition is given below. 0 20 40 60 80 100 0.0 0.1 0.2 0.3 0.4 0.5 0.6

l Useful RA - The ACAS II system generated an advisory in accordance with its technical specifications in a situation where there was or would have been a risk of collision between the aircraft. l Nuisance RA - The TCAS II system generated an advisory in accordance with its technical specification in a situation where there was not, or would have not been, a risk of collision between the aircraft. l Unclassifiable RA - The TCAS II system generated an advisory that cannot be classified because of insufficient data.

2009 The extent to which pilots and controllers correctly follow these classifications when filling out their reports is unknown. The2008 assessment of events (e.g. severity, usefulness) is based on subjective assessment of those who were involved. Nevertheless, these values2007 arguably 2006 2009 provide a general measure of the current operational acceptability of ACAS. 2008

0 10 20 30 40 50 60 0.0 0.1 0.2 0.3 0.4 0.5

Figure 28: ACAS RA Classification. 2009 Unclassifiable RA 2009 2008 Summer 2008 - 2009 per 10,000 2008 2007 operations 2006 Nuisance The majority of the RAs that occurred during summer 2009 have been classified Useful RA by pilots as Useful RAs. 0 20 40 60 80 100 0.0 0.1 0.2 0.3 0.4 0.5 0.6

EVAIR Summer Seasons 2006-2009 Evolution 21 EUROCONTROL

EVAIR_bulletin04.indd 21 2009 15/02/10 14:15 2008 2007 2006 2009 2008

0 10 20 30 40 50 60 0.0 0.1 0.2 0.3 0.4 0.5 2009 2009 2008 2008 2007 2006

RA geometry between two aircraft 0 20 40 60 80 100 0.0 0.1 0.2 0.3 0.4 0.5 0.6

Figure 29: ACAS RA Geometry. Summer 2006 - 2009 (absolute figures) Head On During summer 2009, the number of ACAS High Vertical Rate incidents triggered by high vertical rate has

decreased significantly in comparison with Horizontal Track Crossings the summer 2008. Vertical Crossings geom- etry between two aircraft increased nine Level - Level times during summer 2009 vs 2008. More Offset 2009 attention will be paid to this trend during 2008 future monitoring. Parallel Arrival 2007 2006 2009 Tail Chase 2008 2009 2009 Vertical Crossings 2008 2008 2007 2006 0 10 20 30 40 50 60 0.0 0.1 0.2 0.3 0.4 0.5

0 20 40 60 80 100 0.0 0.1 0.2 0.3 0.4 0.5 0.6

Figure 30: ACAS RA Geometry. Head On Summer 2008 - 2009 per 10,000 flight operations High Vertical Rate During summer 2009, Vertical crossings RA geometry increased significantly having Horizontal Track Crossings more than 0.2 incidents per 10,000 flights. It is encouraging that during summer 2009 Level - Level High Vertical Rate as one2009 of the causals for the activation of the RA saw a two-fold 2008 Offset reduction compared with2007 summer 2008. 2006 2009 Parallel Arrival 2008

Vertical Crossings

0 10 20 30 40 50 60 0.0 0.1 0.2 0.3 0.4 0.5

EUROCONTROL 22 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 22 15/02/10 14:15 RA instructions

Figure 31: ACAS RA Instructions. Summer 2006-2009 (absolute figures) Climb RA Reduce/Adjust RA continues to show an increasing trend and has the largest Crossing RA number of incidents during summer 2009. Descend RA The number of Climb RAs and Monitor Vertical Speed increased during summer Increase RA 2009. 2009 2009 2008 2008 Maintain vertical speed RA 2007 2006 Monitor vertical speed RA

Reduce/Adjust RA 0 10 20 30 40 50 60 0.00 0.05 0.10 0.15 0.20 0.25 0.30

18 16 2009 14 2008 Figure 32: ACAS RA Instructions. Climb RA 2007 Summer 2008-2009 per 10,000 flight 12 2006 operations 10 Crossing RA It should be noted that ICAO no longer 8 requires the collection of RA data when there Descend RA 6 is no contradiction with an ATC clearance. 4 Increase RA 2009 2 2009 2008 2008 0 2007 Maintain vertical speed RA 60 70 80 90 98 2006 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 410 Monitor vertical speed RA

Reduce/Adjust RA 0 10 20 30 40 50 60 0.00 0.05 0.10 0.15 0.20 0.25 0.30

300

250 2009 18 2008 16 2007 2009 200 14 2008 150 12 2007 2006 10 EVAIR Summer Seasons 2006-2009 Evolution 23 100 EUROCONTROL 8 50 6 4 EVAIR_bulletin04.indd 23 0 15/02/10 14:15 0 10 20 30 40 50 60 70 80 90 2 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 400 410 0 60 70 80 90 98 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 410

300

250 2009 2008 200 2007

150

100

50

0 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 400 410 2009 2009 2008 2008 2007 2006

0 10 20 30 40 50 60 0.00 0.05 0.10 0.15 0.20 0.25 0.30

ACAS FL distribution

18 16 2009 14 2008 12 2007 2006 10 8 6 4 2 0 60 70 80 90 98 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 410

Figure 33: ACAS Flight Level Distribution comparison Summer Seasons 2006-2009.

Statistics related to the FL distribution of the manually collected ACAS incidents show that the ACAS incidents are located in clusters of FLs. During summer 2009 a large number of FLs show a significant increase in the number of ACAS incidents. Examples are: 70 - 150; 180 - 240; 290 - 380. 300

RA horizontal relative movement 250 2009 2008 200Figure 34: ACAS Horizontal Relative 2007 Movement. Crossing track 2009 2009 2008 2008 150Summer 2006 - 2009 (absolute figures) Diverging tracks 2007 During summer 2009 there was a drastic 2006 100reduction of all type of the Horizontal Rela- Parallel Track

tive Movement ACAS incidents. As it has Reciprocal tracks been50 already shown at Figures 27 and 29 the largest number of ACAS incidents occurs Same track 0

during Vertical0 change of the altitudes. 0 5 10 15 20 25 30 35 40 0.00 0.05 0.10 0.15 0.20 0.25 0.30 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 400 410

Crossing track Figure 35: ACAS Horizontal Relative 2009 2009 Movements. Summer 2008 - 2009 per2008 18 2008 Diverging tracks 2007 10,000 flight operations 16 2006 2009 During summer 2009 all type of the Hori- Parallel Track 14 2008 zontal Relative Movements show a signifi- Reciprocal tracks 12 2007 cant decrease. Crossing tracks which have 2006 10 the highest level in 2009 have less than Same track 8 0 0.035 incidents10 per15 10,00020 flights.25 30 35 40 0.00 0.05 0.10 0.15 0.20 0.25 0.30 6 4 2 0 18 60 70 80 90 98 EUROCONTROL 24 EVAIR100 Summer110 120 Seasons130 140 150 2006-2009160 170 180 Evolution190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 410 16 2009 14 2008 12 2007 2006 EVAIR_bulletin04.indd10 24 15/02/10 14:15 8 6 4 2 0 60 70 80 90 98 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 410 PART TWO - ACAS RAs COLLECTED AUTOMATICALLY FROM ONE MODE-S RADAR STATION

The Automated Safety Monitoring Tool (ASMT) is being used to record and analyse all TCAS advisory messages down linked by one Mode S radar in European airspace. This set of statistics has been assembled from data that has been collected between 01 April and 30 September in 2007, 2008 and 2009.

Number of events recorded

Events with valid Events with erroneous Complex Year Totals RA messages RA messages RA Events

2007 1032 6289 25 7346

2008 937 3393 24 4354

2009 668 3032 19 3719

Events with valid RA messages = RA downlink messages which are not empty or do not contain only the stop bit. Events with erroneous RA messages = RA downlink messages which are empty or contain only a stop bit. Complex RA events = Events where more than two different RAs are detected for one aircraft during the encounter.

The figures show a significant drop in the number of RAs over the three summer periods, i.e. a drop of 9% in 2008 and a further drop of 29% in 2009.

During 2007 the monitoring was not 24/7 due to radar maintenance and change of ASMT monitoring machine. In 2008 automatic filtering of erroneous messages from selected airframes was also introduced. In 2009 automatic monitoring was transferred from monitoring one radar on line to processing recorded radar LAN data from 10 radars. The data presented here covers one radar, to enable comparison of statistics.

Downlink anomalies

In summer 2009 there were 110 airframes detected that incorrectly issued downlink requests, and 20 were detected 10 or more times during the 6 month period. From the 3032 Erroneous Message events, 2258 were from this set of 20 airframes; and 2101 of these erroneous messages were from 12 airframes.

In summary, a small number of aircraft are non-compliant with the SARPS and ‘pollute’ the RF environment. Corrective action is under- way with the operators and manufacturers concerned.

EVAIR Summer Seasons 2006-2009 Evolution 25 EUROCONTROL

EVAIR_bulletin04.indd 25 15/02/10 14:15 2009 2009 2008 2008 2007 2006

0 10 20 30 40 50 60 0.00 0.05 0.10 0.15 0.20 0.25 0.30

18 16 2009 14 2008 12 2007 2006 10 8 6 4 2

RA0 distribution by flight levels 60 70 80 90 98 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 410

Figure 36 below shows the number of RAs by Flight Level and compares 2007, 2008 and 2009. All recorded RAs with a valid downlink message (not blank and not stop only) have been counted even if the intruding aircraft has not been identified. FLs in this report are rounded to the nearest FL. There is no major trend difference between the 2007, 2008 and 2009 distributions except for a reduction in higher FL RAs.

300

250 2009 2008 200 2007

150

100

50

0 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 400 410

Figure 36: Advisories by Flight Level

There is a large proportion of RAs between FL10 and 30. They are mostly the result of VFR traffic in uncontrolled airspace. (Figure 37 over leaf show the high number of Mode C Intruders at these levels, but note the increase in Mode S equipped intruders in 2008 compared with 2007 indicating a greater number of GA aircraft fitted with Mode S transponders).

EUROCONTROL 26 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 26 15/02/10 14:15 Intruder equipage by flight bands

The diagram below shows the number of RA events 2007 Below 3000 recorded on both aircraft and the number of events 2008 Below 3000 2009 Below 3000 where only one aircraft reported an RA. 2007 FL 30-100 2008 FL 30-100 Events shown as an RA with a Mode S intruder are 2009 FL 30-100 to a very large extent TCAS equipped. TCAS does 2007 FL 100-180 2008 FL 100-180 not symmetrically generate RAs. There are many 2009 FL 100-180 events where only one aircraft receives an RA, even 2007 FL 180-290 though both are TCAS equipped. RA intruder Mode 2008 FL 180-290 Both RA C are confirmed as Mode C intruders by the Own RA 2009 FL 180-290 Intruder ModeS 2007 Above FL 290 downlink. In the lower levels there is an increased Intruder Mode C 2008 AboveFL 290 percentage of aircraft identified as Mode S intruders 2009 AboveFL 290 and a reduction in Mode C intruders which may be 0 100 200 300 400 500 600 due to the increase in the number of General Aviation aircraft fitting Mode S transponders. Figure 37: TCAS equipage in Encounters

Both RA Note: The number of Intruderevents ModeS at each level is lower than the number of RAs because more than one RA can exist for each event. Intruder Mode C

0 At100 lower200 flight300 levels,400 the 500majority600 of RAs are against Mode C intruders, whereas at higher levels, most RAs are against intruders with Mode S or are coordinated TCAS encounters. A significant reduction of Level Off RAs in upper flight levels is possibly due to the increased use of 25ft reporting Mode S transponders.

200 Hot Spots

200 Plotting XY distributions of RAs at different altitude 150 ranges allows “RA hotspots” to be identified. Frequently 150 these are related to airspace design and airspace classi- fication issues, e.g. where climbing aircraft level off just 100 100 below a flight level where descending aircraft level off

50 too. As an example, a hot spot diagram is shown below. A clearer conclusion could been given if the underlying 50

0 route network had been shown; however, to preserve -200 -150 -100 -50 0 50 100 150 200 confidentiality this is not possible. The circled regions -50 correspond to stacks and sector boundaries. 0 -200 -150 -100 -50 0 50 100 150 200

-100 -50

-150

-200 Figure 38: Example of Hot Spot Diagram using summer 2009 data -100

-150

-200 EVAIR Summer Seasons 2006-2009 Evolution 27 EUROCONTROL

EVAIR_bulletin04.indd 27 15/02/10 14:15 Pilot response 100%

90% For an2008 RA that does not change for at least 2 radar cycles, i.e. 10 seconds (536 events - 60%), we see that 93% have responded to the RA. There2007 were 76% achieving the requested vertical rate, 4% exceeding the requested rate and 12% either slow or failing to achieve the 80% requested rate; 8% were calculated to give opposite responses 70%

60% Opposite Excessive 50%100% 8% 4%

40%90% 2008 2007 Below Followed 30%80% requested rate 76% 12% 20%70% 69% 27% 4% 10%60% 58% 29% 13% 0%50% Normal High Level 40%

30%

20% 69% 27% 4% 10% 58% 29% 13% 0% Figure 39: Response to Standard RAs Normal High Level

900 Opposite Excessive 811 800 29% 4%

731 Followed 700 26% 655 2008 600 2007

500 468 900 419 400 811 800 351

300 731 700 236 655 2008 200 192 600 2007 100 Below 500 33 468 27 13 13 13 20 requested rate 10 0 9 8 1 9 5 6 0 419 41% 400MVS KVS AVS CL ICL CCL RCL DE IDE CDE RDE 351 300 Figure 40: Response to Changing RAs 236 200 192

100

33 27 13 13 13 20 0 10 0 9 8 1 9 5 6 MVS KVS AVS CL ICL CCL RCL DE IDE CDE RDE For Changing RAs (353 events - 39%), where the RA is observed to change after one radar cycle (changes in less than 10 seconds), compliance with the RA is not as good, even though TCAS expects a quicker response from pilots. The high level of opposite responses to changing RAs is a cause for concern and re-emphasises the need for pilots to be trained to follow the full sequence of RAs that they receive.

EUROCONTROL 28 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 28 15/02/10 14:15 Figure 41 below shows responses to all type of RAs for 2007, 2008 and 2009. It can be seen that there is a steady decrease in the numbers of Not Followed, but an increase in those going Opposite to the direction requested.

60% 2009 2008 50% 2007 2009 40%60% 2008 50% 30% 2007 20%40%

10%30% 2007 2008 20%0% 2009 10% Excessive Followed NotFollowed Opposite 2007 2008 0% 2009 Figure 41: ComparisonExcessive toFollowed all type RANotFollowed responses overOpposite period 2007 - 2009

Vertical rates at the time of the RA

At the time of the first RA reports, analysis evaluation show that in 2009 18% of aircraft had a high vertical rate exceeding 1500fpm, which is 5% less than in 2008. This indicates an increasing compliance with the PANS OPS recommendation overleaf which appears to have resulted in a reduction in the number of RAs in the observed airspace.

900

800

700900

600800

500700

400600

300500

200400

100300 Level 200 Normal 0 High 100 2007 2008 2009 Level Normal 0 High 2007 2008 2009

Figure 42

EVAIR Summer Seasons 2006-2009 Evolution 29 EUROCONTROL

EVAIR_bulletin04.indd 29 15/02/10 14:15 2007 2008 2009

High 26% 23% 18%

Normal 50% 63% 62%

Level 24% 14% 20%

Pilots should be aware of the following provision in PANS-OPS that is in force from 20th November 2008:

“Pilots should use appropriate procedures by which an aeroplane climbing or descending to an assigned altitude or flight level, especially with an autopilot engaged, may do so at a rate less than 8 m/s (or 1 500 ft/min) throughout the last 300 m (or 1 000 ft) of climb or descent to the assigned altitude or flight level when the pilot is made aware of another aircraft at or approaching an adjacent altitude or flight level, unless otherwise instructed by ATC. These procedures are intended to avoid unnecessary airborne collision avoidance system (ACAS II) resolution advisories in aircraft at or approaching adjacent altitudes or flight levels. For commercial operations, these procedures should be specified by the operator.”

EUROCONTROL 30 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 30 15/02/10 14:15 Advisories issued

The number of RAs is greater than the number of events for two reasons: more than one RA can be issued to an aircraft during an event; and both own aircraft and intruder can issue an RA. The table below shows the breakdown of advisories issued in the collected data.

The reduction in the number and percentage of AVSA RAs is probably due to the recommendation to reduce vertical rate below 1500fpm in the last 1000ft when approaching a cleared level.

Code Advisory 2007 2008 2009 AVS Adjust Vertical Speed 42.2% 40.5% 34.0% MVS Monitor Vertical Speed 21.4% 22.4% 24.0% CL Climb 19.9% 22.1% 22.0% DE Descent 11.8% 11.1% 12.4% KVS Maintain Vertical Speed 2.0% 1.7% 2.0% ICL Increase Climb 0.7% 0.6% 2.4% ICD Increase Decent 0.6% 0.7% 1.4% RCL Reversal Climb 0.2% 0.6% 0.5% RDE Reversal Descent 0.3% 0.0% 0.0% CCL Crossing Climb 0.3% 0.0% 0.0% CDE Crossing Descent 0.7% 0.2% 0.0%

The TCAS advisories AVS and MVS do not require deviation from ATC clearance unless ATC requests a specific vertical rate. They correspond to 58% of RAs in 2009 representing a drop of 5.5% from 2007.

500 491

450 454 400 2009 357 350 2008 2007 300 231 248 231 250 249 251 252 200

150 137 124 130 100

50 23 19 21 8 7 25 4 2 7 5 7 8 15 8 2 3 15 0 MVS KVS AVS CL ICL CCL RCL DE IDE CDE RDE Figure 43: Distribution by RA Type

07:23:30 07:24:00 07:24:30 07:25:00 07:25:30 EVAIR Summer Seasons 2006-2009-48 Nm Evolution GLF3 31 EUROCONTROL

-50 Nm COC TA FL350 LC1 COC -52 Nm 348 ft EVAIR_bulletin04.indd 31 LC5 15/02/10 14:15 IDE A320 TA CL ICL -54 Nm IDE FL340 IDECOC ICL CL 0.9 Nm TA COC DE -56 Nm LC5 A320 FL330 -58 Nm LC1

TA -60 Nm FL320

-62 Nm GLF3

36 Nm 38 Nm 40 Nm 42 Nm 44 Nm 46 Nm 48 Nm 50 Nm 52 Nm FL310

LC5 500 491

450Adjust vertical speed454 400 Interesting AVS event from 2009 data 2009 357 350 2008 2007 300Analysis of a serious separation loss, resulting from misinterpretation of TCAS Resolution Advisory to Adjust Vertical Speed (always means 231 248 231 250decrease249 251 252 vertical speed) is shown below. This is a well documented problem and changes have been incorporated into TCAS Version 7.1 200altering the Advisory annunciation to “Level Off Level Off”. If V7.1had been fitted and followed in this case, GLF3 would have levelled at the cleared level instead of Climbing at an increased rate. 150 137 124 130 100 EVAIR TCAS simulation based on the radar data, shows that GLF3 was climbing at 2500fpm to cross the path ahead of A320 which was maintaining50 FL340. The strength of the first Advisory to GLF3 cannot be confirmed from downlinked data but the downlinked message 23 19 21 8 7 25 4 2 7 5 7 8 15 8 2 3 15 showed0 that this was a limit climb which could have allowed a climb of up to 2000fpm (although simulation suggested the limit was MVS KVS AVS CL ICL CCL RCL DE IDE CDE RDE 1000fpm). As the climb of GLF3, after the adjust vertical speed RA, was increasing above (2500fpm), the advisory to adjust vertical speed could have been misunderstood. Note: ACAS Safety Bulletin No 3 addresses the issue of misunderstanding Adjust vertical speed RA’s (see link below).

07:23:30 07:24:00 07:24:30 07:25:00 07:25:30 -48 Nm GLF3

-50 Nm COC TA FL350 LC1 COC -52 Nm 348 ft LC5 IDE A320 TA CL ICL -54 Nm IDE FL340 IDECOC ICL CL 0.9 Nm TA COC DE -56 Nm LC5 A320 FL330 -58 Nm LC1

TA -60 Nm FL320

-62 Nm GLF3

36 Nm 38 Nm 40 Nm 42 Nm 44 Nm 46 Nm 48 Nm 50 Nm 52 Nm FL310

Figure 44

LC5

EUROCONTROL 32 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 32 15/02/10 14:15 SECTION 5 - MISCELLANEOUS

In this section of the Safety Bulletin we take a look at EVAIR data received on ‘third party’ incidents that are not directly attributable to ATC but which can have serious aviation safety consequences. The ‘third party’ data includes:

l Bird strikes

l Vehicles, Person and Animals (as they affect airfield operations).

l Lasers (improper illumination of aircraft and ATC towers).

In this Bulletin we take a closer look at the growing menace of the improper use of lasers pointers.

LASERS

EVAR Bulletin No 3 highlighted the growing menace to aviation caused by the misuse of hand-held laser devices which are being used to ‘target’ aircraft and sometimes ATC facilities.

Regrettably, data gathered through the EVAIR process since publication of the last EVAIR Bulletin confirms that this threat is widespread across Europe. There are no universal solutions but some countries have been strengthening their legislative frameworks to counter the menace. This has led to a number of successful prosecutions against the perpetrators in some countries which will hopefully act as a deterrent.

EVAIR will continue to monitor the situation and welcomes additional reports concerning the misuse of lasers against aviation assets. Further action will be considered if it is appropriate. In the meantime, an article about lasers and aviation is available on SKYbrary (www.skybrary.aero). It contains links and references to the mass of material available on the topic and is a good place to start research into this aviation hazard.

2008 6

2007 1 Figure 45: Laser. Summer 2007- 2009 After publication of the EVAIR Safety Bulletin 3 and the request for more reports, the number of reported Laser incidents significantly increased in summer 2009 and from one attack reported in 2007 we have 29 of them reported during summer 2009.

2009 29

EVAIR Summer Seasons 2006-2009 Evolution 33 EUROCONTROL

EVAIR_bulletin04.indd 33 15/02/10 14:15

Crossing Crossing Limit Limit

Crossing

Chasing -Head-On Obtuse Acute Limit Limit

Chasing Head-On

Chasing -Head-On Obtuse Acute Limit Limit Crossing

Crossing Crossing Limit Limit

2008 (0,043) 22%

Figure 46: Laser. Summer 2008 – 2009 per 10,000 flight operations The number of Laser incidents in 2009 had higher rate per 10,000 operations than Level Busts and Runway Incursions together. It indicates that action is required to combat this growing threat to aviation safety.

2009 (0,15) 78%

Crossing Crossing Limit Limit

Crossing

Chasing -Head-On Obtuse Acute Limit Limit

Chasing Head-On

Chasing -Head-On Obtuse Acute Limit Limit Crossing

Crossing Crossing Limit Limit

EUROCONTROL 34 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 34 15/02/10 14:15 EVAIR AIRLINES ’ CORNER

INTRODUCTION

As one of the EVAIR data providers, Brussels Airlines is happy to contribute to this new Safety Bulletin, highlighting some areas which could be widespread across Europe.

The data provided could be compared with trends identified by EVAIR on the same field. The following article gives a statistical analysis of the ATM incident contributions along with Runway Incursion, Level Bust and Airspace Infringement trends and the main causal factors from our perspective.

OPERATIONS

Brussels Airlines Fleet BAe146-200 6 Figure 47: Brussels Airlines Fleet AvroRJ 85 14 To put the figures in this article in perspective and as a matter of intro- AvroRJ100 12 duction to Brussels Airlines, (Figure 47) represents the company fleet. B737-300 5 There are about 3000 employees and 51 aircraft that operate daily B737-400 6 some 300 flights. The operations are divided into Short & Medium Haul A319 4 and Long Haul connecting the ‘Capital of Europe’ to some 70 premium A330 4 European and African airports. Total 51

EVAIR Summer Seasons 2006-2009 Evolution 35 EUROCONTROL

EVAIR_bulletin04.indd 35 15/02/10 14:15 SAFETY MANAGEMENT SYSTEM – BRUSSELS AIRLINES

Senior Management Conformity Validation Committee Corrective Action Plan Non compliance Accountable Manager All Executive Functions Rejected Quality Control Non Acceptable Safety Goals ● Process validation ● Procedures control & Risk Level Acceptable Risk ● (AOC, OM, FCOM, IOSA) Control Owner of the Risk Monitor Knowledge transfer ● Eliminate Corporate Quality Assurance & Record & Awareness ● Mitigate Risk Database ● ● Identify Systematic De„ciencies Aknowledgement Tolerate ● Audit (Special, Periodic, Annual) & Feedback ● Quality Awareness Follow up Rejected Follow-up V & Record & Analysis & Risk Corporate Operational Assesment Closure Risk Identi cation Pro-active & Reactive Risk Identi„cation REPORTED Investigation Air Safety Report Risk Classi„cation Flight Crew Report Recommendation Cabin Crew Report Awareness Maintenance & Engineering Report Author/ Source Ground Handling Report Human Factors Report Fatigue Report OBERVED Line Checks Observer ights Audits RECORDED Flight Data Monitoring Engine Trend Monitoring

Figure 48: Brussels Airlines Safety Management System

The chart above depicts the Brussels Airlines Safety Management System with its Risk Identification and Corrective Action Plan. Ideally this would be a closed loop system, which is the case when all parties involved are linked in first degree to the organisation. However, when actions are required from third parties, this loop has a loose endIncident when classi cation it comes to the corrective action plan, as indicated on the schematic representation above, it gets much more difficult to assure proper feedback and follow-up as it is not controlled within the organisation itself but in fact outsourced to the third party (for instance ANSPs).

Descriptors Factors What concerns the decision making process on the proposed corrective actions within the airline is the responsibility and accountability of the Senior Management. Whenever it is an ATM incident for instance, it is entirely up to the air navigation service providers to close the loop. Immediate E‚ects Event Type Factors - Major

This is an aera where EVAIR can become a major facilitator in providing feedback to both sides and act as the missing link between the airlines and the service providers. To close the loop is of theDescriptors utmost importance to satisfactorily carry out the post analysisFactors and take - Minor correc- tive measures, meeting the standards we are all striving to achieve.

EUROCONTROL 36 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 36 15/02/10 14:15 Senior Management Conformity Validation Committee Corrective Action Plan Non compliance Accountable Manager All Executive Functions Rejected Quality Control Non Acceptable Safety Goals ● Process validation ● Procedures control & Risk Level Acceptable Risk ● (AOC, OM, FCOM, IOSA) Control Owner of the Risk Monitor Knowledge transfer ● Eliminate Corporate Quality Assurance & Record & Awareness ● Mitigate Risk Database ● ● Identify Systematic De„ciencies Aknowledgement Tolerate ● Audit (Special, Periodic, Annual) & Feedback ● Quality Awareness Follow up Rejected Follow-up V & Record & Analysis & Risk Corporate Operational Assesment Closure Risk Identi cation Pro-active & Reactive Risk Identi„cation REPORTED Investigation Air Safety Report Risk Classi„cation Flight Crew Report Recommendation Cabin Crew Report Awareness Maintenance & Engineering Report Author/ Source Ground Handling Report Human Factors Report Fatigue Report OBERVED Line Checks Observer ights Audits RECORDED Flight Data Monitoring Reports analysed within Brussels Airlines Engine Trend Monitoring Yearly about 63,000 flights generate more than 800 Air Safety Reports (ASRs), giving an overall ratio of about 13 ASRs/1000 flights over the years, and still steadily increasing. All ASRs – along with 6 other report types, as indicated in Figure 49– are put into Sentinel (our Safety & Incident Management System) and are given a risk rating and classification1. This classification encloses the determination of the Immediate Effects, Event Types & Descriptors for what concerns the event itself, and the contributory Factors, leading to the incident. It is by such classification systems that trend analysis can be started and results in some of the statistics that are discussed further on.

Incident classi cation

Descriptors Factors

Immediate E‚ects Event Type Factors - Major

Descriptors Factors - Minor

Figure 49: Incident Classification

SOME STATISTICS

An overview of the total number of ASRs (until November 2009) in terms of absolute as well as relative figures is given in Fig 50; the percentage indicated in the table represents the contribution of the respective topic with regards to the total number of reports.

Number of ASRs / ATM Level busts RWY RWY Inc Year ATM Level busts ASRs 1000 flights Percentage Percentage Incursions Percentage

2003 318 - - - 20 6,3% 8 2,5%

2004 517 - - - 37 7,2% 2 0,4%

2005 795 - 41 5,2% 1 0,1% 4 0,50%

2006 838 - 58 6,9% 9 1,1% 3 0,36%

2007 841 13,79 60 7,1% 8 1,0% 2 0,24%

2008 871 12,73 54 6,2% 15 1,7% 2 0,23%

2009 663 12,55 39 5,9% 3 0,5% 2 0,30%

Figure 50: Overview of absolute and relative values of ATM related reports

1 The Descriptor classification system within Sentinel is an industry standard and is used by IATA for their STEADES program. It was developed in conjunction with IATA and is the Intellectual property of Mercator.

In the Risk area the ECCAIR risk assessment tool is integrated with the Airbus Risk Matrix (severity versus reoccurrence) and aims to provide guidance to the risk assessment.

EVAIR Summer Seasons 2006-2009 Evolution 37 EUROCONTROL

EVAIR_bulletin04.indd 37 15/02/10 14:15 AIR TRAFFIC MANAGEMENT

2008 60% 2008 2007 60% 2007 50% 2006 50% 2006 2005 40% 2005 2004 40% 2004 30% 2003 30% 2003 20% 20% 10% 2007 10% 20072005 0% 20052003 0% 2003 Other Airprox Other Airprox ATC English ATC Groundprox ATC English ATC Groundprox Met Info/Brie ng Callsign Confusion Met Info/Brie ng ATC Service Standard ATC Callsign Confusion Comms with ATC Lost with ATC Comms Excessive Hold Delays Hold Excessive Inadequate Separation Inadequate ATC Service Standard ATC Comms with ATC Lost with ATC Comms Excessive Hold Delays Hold Excessive Inadequate Separation Inadequate Take O€ Clearance Cancelled Take Take O€ Clearance Cancelled Take Wake Turbulence - Encountered Turbulence Wake Landing Clearance Received Not Wake Turbulence - Encountered Turbulence Wake Landing Clearance Received Not Landing Clearance with Runway In Use Landing Clearance with Runway Landing Clearance with Runway In Use Landing Clearance with Runway Figure 51: Air traffic management

ATC Service Standard 33% ATC Service Standard 2% Wake Turbulence - Encountered 2% Wake TurbulenceComms - Encountered with ATC Lost 2% Comms with ATC Lost 2% Inadequate Separation 2% Inadequate AirproxSeparation 2% Airprox Landing Clearance with Runway In Use Landing ClearanceExcessive with Hold Runway Delays In Use Excessive HoldCallsign Delays Confusion Callsign ConfusionTake O€ Clearance Cancelled Take O€ ClearanceMet Info/Brie ng Cancelled Met Info/Brie ngLanding Clearance Not Received 3% Landing ClearanceATC English Not Received 3% ATC English Groundprox 3% Groundprox Other 4% 16% Other 8% 14%

Figure 52: ATM events - Different Event Types involved

Over the years “ATC service standard” has been the major contributor to Air Traffic Management events and is on top of the ranking. Considering the absolute values, it is clear that the number of these related reports are increasing (6 in 2003 to 26 in 2008). Note however, there always has been a steady increase in general reporting. When making this issue relative to the number of reported ATM events, it increased from 23% in 2003 to 38% in 2008. Unfortunately we lack profound historical data, to be able to consider trends over the years. However when considering Figure 50 it might be reckoned that there is a little increase in ATM reported events. Most of the ATC service standard issues are related to “confusing ATC clearances”, “poor ATC” and their contribution to some air misses or TCAS RA’s.

EUROCONTROL 38 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 38 15/02/10 14:15 2008 60% 2007 50% 2006 2005 40% 2004 30% 2003

20%

10% 2007 2005 0% 2003 Other Airprox ATC English ATC Groundprox Met Info/Brie ng Callsign Confusion ATC Service Standard ATC Comms with ATC Lost with ATC Comms Excessive Hold Delays Hold Excessive Inadequate Separation Inadequate Take O€ Clearance Cancelled Take Wake Turbulence - Encountered Turbulence Wake Landing Clearance Received Not

Encountered wake turbulence In Use Landing Clearance with Runway has been stable over the years and count for approximately 16% of the total ATM events, although in almost every case the required separation minima were met.

The third most common ATM event is the Loss of Communication. The great majority of these events are due to Human Factors: honest ATC Service Standard mistakes made by the flight crew (attention failure, vigilance/failure of crosschecking, flight crew miss selection or aural misperception). All Wake Turbulence - Encountered these humanComms withfactors ATC Lost errors are internally classified using the Human Factors Analysis and Classification System © by Wiegmann & Shapell. A minorityInadequate of these Separation events are related to technical malfunctions. Airprox Landing Clearance with Runway In Use ComparingExcessive the Hold European Delays and African operations. African flights represent 5% of the total number of flights and generate 4% of the total Callsign Confusion ATM related reports. Take O€ Clearance Cancelled Met Info/Brie ng Landing Clearance Not Received ATC English Groundprox LEVEL BUSTSOther

Technical Flight Management 10% 30%

Weather 30% ATM 30%

Figure 53: Level busts – Different Event Types involved

RUNWAY INCURSION

During the past 5 years (2005-2009) 13 active Runway Incursions were reported as shown in Figure 50. Because of this little number (luckily after all), no credible trend analysis could be made over the years.

As this is a Flight Management issue, it is obvious that in all cases, Human Factors of the Flight Crew were involved. However, in 4 incidents 2 major contributory factors were Airport Management (RWY Markings/Lightings) and ATM (ATC violations and communication error).

No hot topic location could be pinpointed since every incident occurred on a different airport. In cooperation with the Local RWY safety team of the airport where Airport Management was a factor, some actions had been taken with regards to the lighting and the use of H24 stop bars. In another instance where ATM was involved, ATC provided recurrent training for the staff involved.

Concerning the flight crew: every incident was analysed and properly debriefed to the crew involved and general comments to the flight crew community were published in the three-monthly Brussels Airlines Safety Magazine, along with the feedback of the third party ANSPs.

EVAIR Summer Seasons 2006-2009 Evolution 39 EUROCONTROL

EVAIR_bulletin04.indd 39 15/02/10 14:15 BRUSSELS AIRLINES SPECIFIC ATM TOPICS

Call sign similarities History

SN Airholding was created in 2002, directly after the bankruptcy of the Belgian national airline Sabena. The new company was called SN Brussels Airlines.

In October 2004, it was decided that SN Brussels Airlines and Virgin Express would join together under the general ownership of SN Airholding, but would each retain their brand and operations independence in the market. Subsequently, in March 2006, the two companies announced the merger and presented to the public a new name – Brussels Airlines.

Operated by DAT

Figure 54: Call sign issues

When SN Brussels Airlines was created out of Delta Air Transport as the successor of Sabena, the call signs used linked the new airline with the former national airline. Because the airplanes were still flying with an “S” (from the era of Sabena) on the tail, the new voice call sign was born: “ESTAIL”.

However, the ATC call sign of Delta Air Transport was DAT.

This missing ‘link’ between what the ATCOs saw on their screens as the ATC call sign and the voice call sign they should be using, just increased the general confusion and raised several call sign issues during regular day-to-day operations. The ATCOs sometimes used the old voice call signs or used the full ATC call sign (eg: “Delta Air Transport”, “Delta Air”,..) or even mixed the two during the same communication. It is obvious that this issue only increased the confusion and several instructions/clearances could have been missed by the flight crew.

Therefore in cooperation with ATS units, Brussels Airlines changed the ATC and Voice call signs several times. However, sometimes these created some confusion with other airlines’ call signs, which compelled us to revert back to the old ones or search for a new one. Luckily after all, we ended up with – currently still without any downside - the current voice call sign BEELINE, linked to the ATC call sign BEL. Hopefully after all these years, the call sign story has finally come to an end.

EUROCONTROL 40 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 40 15/02/10 14:15 MAIN STAKEHOLDERS WORKING TOGETHER TO IMPROVE ATM SAFETY

Through trend analysis of the Air Safety Reports, begining in 2005, Brussels Airlines started to detect an ATM Safety Issue at Madrid. The reports tackled a wide variety of issues on ground as well as in the air.

What follows is a chronological list of actions carried out by Brussels Airlines and the consequent escalation by different other international players in Flight Safety:

10 March 2005, Madrid: IATA awareness raising visit on RWY incursion prevention;

7 July 2005: Brussels Airlines seeking assistance with IATA, AEA & other operators;

22 December 2005, Madrid: IATA high-level meeting on MAD safety issues;

15-16 April 2008, Madrid: Workshop Eurocontrol - improving Madrid Barajas operational and flight safety performance;

30 Jan 2009: Eurocontrol (EVAIR)/AENA cooperation and provision of the feedback on the ATM incidents. Thanks to the cooperation with all these different actors, the number of reports related to ATM Madrid decreased tremendously after 2006 to end up with no more reported issues in 2009.

EVAIR Summer Seasons 2006-2009 Evolution 41 EUROCONTROL

EVAIR_bulletin04.indd 41 15/02/10 14:15 ANNEX 1 - EUROPEAN ACTION PLANS

THE EUROPEAN ACTION PLAN FOR THE PREVENTION OF LEVEL BUST

Reducing Level busts is one of EUROCONTROL’s highest priorities. EUROCONTROL began raising awareness of the Level Bust issue in 2001, organised series of workshops, and established a Level Bust Task Force to define the recommendations and to formulate an action plan to reduce level busts.

The level bust action plan is the product of work carried out by EUROCONTROL’s cross-industry Level Bust Task Force, which was set up in 2003. The Task Force reviewed the evidence available, identified the principal causal factors, and listened to the Air Navigation Service Providers and aircraft operators with experience in reducing level busts.

The Action Plan contains recommendations for Air Traffic Management, Air Traffic Controllers, and Aircraft Operators. It designed to reduce the frequency of level busts and reduce the risks associated with level busts. Implementation of the Action Plan will be monitored by the Task Force monitoring group reporting to the EUROCONTROL Safety Improvement Sub Group (SISG).

http://www.eurocontrol.int/safety/public/standard_page/Level_bust.html

THE EUROPEAN ACTION PLAN FOR THE PREVENTION OF RUNWAY INCURSIONS (EAPRI)

Although runway safety includes issues such as foreign objects, debris and wildlife straying onto the runway and other technical deficien- cies, this action plan specifically addresses the subject of runway incursion prevention.

EAPRI is the result of the combined efforts of organisations representing all areas of aerodrome operations that are totally committed to enhancing the safety of runway operations by advocating the implementation of the recommendations that it contains in the ECAC area. The ICAO secretariat has lent its strong support to the work of this group and urges all States to fully implement the ICAO provisions relevant to runway safety.

The 56 recommendations it contains, when implemented, will enhance runway safety by the consistent and harmonised application of existing ICAO provisions, improving controller - pilot - vehicle driver communications and working procedures at the aerodrome, and by the subsequent increase in situational awareness.

http://www.eurocontrol.int/runwaysafety/public/subsite_homepage/homepage.html

CALL SIGN SIMILARITY (CSS)

The European Action Plan for Air Ground Communication Safety (conceived inter alia by EUROCONTROL, aircraft operators (AOs) and the Flight Safety Foundation) identified callsign similarity (CSS) as a significant contributor to air-ground communication issues. Analysis of ATC reported events shows that 5% involve incidences where CSS is involved. Some aircraft operators are trying to find solutions; the only known ANSP actively operating a service to de-conflict call signs is France’s DSNA.

Research and CBA studies show that the most cost efficient way of providing a long-lasting, Europe-wide solution is to create a central management service to de-conflict ATC call signs. This strategy provides economies of scale and rapid pay back of investment (3 years). More importantly, it is calculated that it will eliminate over 80% of the CSS incidences and thus improve safety.

http://www.eurocontrol.int/safety/public/standard_page/Callsign_Similarity_project.html

EUROCONTROL 42 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 42 15/02/10 14:15 ANNEX 2 - DEFINITIONS

The following definitions are extracted from the HEIDI and/or HERA Taxonomies. HEIDI (Harmonisation of European Incident Definitions Initiative for ATM) intends to finalise a harmonised set of definitions (taxonomy) for ATM related occurrences. HERA (Human Error in European Air Traffic Management) develops a detailed methodology for analysing human errors in ATM, including all error forms and their causal, contributory and compounding factors.

More information can be found at: HEIDI: http://www.eurocontrol.int/src/public/standard_page/esarr2_heidi.html HERA: http://www.eurocontrol.int/humanfactors/public/site_preferences/display_library_list_public.html#5

DEFINITIONS

ATC clearance/instruction (HEIDI): Related to incorrect or wrong aircraft action. Authorisation for an aircraft to proceed under conditions specified by an air traffic control unit and deviations from the clearance which cause runway incursions, taxiway incursions, apron incur- sions, level bust, unauthorised penetration of airspace etc.

Coordination (HEIDI): internal coordination encompassing coordination with sectors within the same unit, and sectors within the ATC suite; external coordination, civil/civil and civil/military; and special coordination, covering expedite clearance, prior permission required, revision and other special coordinations.

Contributory factors (HEIDI): A part of the chain of events or combination of events which has played a role in the occurrence (either by easing its emergence or by aggravating the consequences thereof) but for which it cannot be determined whether its non existence would have changed the course of events.

Decision-Making (HERA): cover incorrect, late or absence of decision.

Failure to Monitor (HERA): failure to monitor people, information or automation.

Judgement (HERA): mainly associated to separation.

Lapses (HEIDI): psychological issues encompassing: Reception of information, Identification of information, Perception of information, Detection, Misunderstanding, Monitoring, Timing, Distraction, Forgetting and Loss of awareness.

Level bust (HEIDI): Any unauthorised vertical deviation of more than 300 feet from an ATC flight clearance Departing from a previously maintained FL, overshooting, undershooting, leveling-off at a different level than cleared level.

Mental/Emotional/Personality issues (HERA): include the following items

l Mental capacity: loss of picture or Safety Awareness; l Confidence in self, in others, in information, in equipment, in automation; l Complacency; l Motivation/Morale; l Attitudes to others; l Personality traits: aggressive, assertive, under-confident, risk taking;

EVAIR Summer Seasons 2006-2009 Evolution 43 EUROCONTROL

EVAIR_bulletin04.indd 43 15/02/10 14:15 l Emotional status: stressed, post incident; l Mis-stored or insufficient learned information; l Planning: insufficient, incorrect or failed; l Recall of information: ailed, inaccurate, rare information, past information; l Violations: routine, exceptional.

Mistakes (HEIDI): psychological issues encompassing: Information wrongly associated, Workload issues, Information not detected, Failure to monitor, Recall of information, Misunderstanding or insufficiently learned information, Judgement, Planning, Decision making, Assumptions and Mindset.

Operational communication (HEIDI): Air-Ground, Ground-Ground and Use of equipment verification testing. Air-Ground communica- tion encompasses hear back omitted, pilots’ read back, standard phraseology, message construction, R/T monitoring including sector frequency monitoring and emergency frequency monitoring, handling of radio communication failure, unlawful radio communications transmission. Ground-Ground communication refers to the standard phraseology, speech techniques, message construction, standard use of equipment like, radio frequency, telephones, intercoms etc.

RA geometry between two Aircraft (ASMT)

Crossing Crossing Limit Limit

Crossing

Chasing -Head-On Obtuse Acute Limit Limit

Chasing Head-On

Chasing -Head-On Obtuse Acute Limit Limit Crossing

Crossing Crossing Limit Limit

Runway Incursion (HEIDI): Any unauthorized presence on a runway of an aircraft, vehicle, person or object that creates a collision hazard or results in a potential loss of separation.

EUROCONTROL 44 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 44 15/02/10 14:15 Spoken communication (HEIDI): human/human communication encompassing air-ground and ground-ground communications but also call sign confusion, noise interference and other spoken information provided in plain language. Air-ground communication refers to language/accent, situation not conveyed by pilots, pilot’s breach of radio telephony (R/T), workload, misunderstanding/misinterpretation, and other pilot problems. Ground-ground communication refers to misunderstanding/misinterpretation, poor/no coordination.

Taxiway Incursion (HEIDI): Any occurrence unauthorized presence on a taxiway of an aircraft, vehicle, person or object that creates a collision hazard or results in a potential loss of separation

Traffic & Airspace problems (HEIDI): there are four set of causal factors under this element

l Traffic load & complexity, encompassing excessive and fluctuating load, unexpected traffic demand, complex mix of traffic, unusual situations (emergency, high risk, other), Abnormal time pressure, underload and call signs confusion; l Airspace problems composed of flights in non controlled and controlled air space, Airspace design characteristics(complexity, changes, other) and temporary sector activities(military, parachuting, volcanic activity, training); l Weather problems such as poor or unpredictable(snow, slush, ice, fog, law cloud, thunderstorm, wind shear); l Pilot problems concerning language, culture and experience aspects.

Traffic Information (HEIDI): essential and local traffic information provided by an air traffic controller to the pilot. Essential information is related to the provision of traffic information containing:

a. direction of flight of aircraft concerned; b. type and wake turbulence category (if relevant) of aircraft concerned; c. cruising level of aircraft concerned; and d. estimated time over the reporting point nearest to where the level will be crossed; or e. relative bearing of the aircraft concerned in terms of the 12-hour clock as well as distance from the conflicting traffic; or f. actual or estimated position of the aircraft concerned.

Local traffic in this context consists of any aircraft, vehicle or personnel on or near the runway to be used, or traffic in the take-off and climb-out area or the final approach area, which may constitute a collision hazard to the other aircraft and about which the information has to be provided.

Workload issues (HERA): concern both minimal and excessive workload

EVAIR Summer Seasons 2006-2009 Evolution 45 EUROCONTROL

EVAIR_bulletin04.indd 45 15/02/10 14:15 ANNEX 3 - ACRONYMS

ACAS Airborne Collision Avoidance System AENA Aeropuertos Españoles y Navegacíon Aéreas (ES) ANSP Air Navigation Services Provider AO Aircraft Operator ASMT ATM Safety Monitoring Tool ATC Air Traffic Control ATM/CNS Air Traffic Management/Communication, Navigation, Surveillance CFMU Central Flow Management Unit CSS Call Sign Similarity ECAC European Civil Aviation Conference EVAIR EUROCONTROL Voluntary ATM Incidents Reporting FL Flight Level GAT General Air Traffic HEIDI Harmonisation of European Incident Definitions Initiative for ATM HERA Human Error in European Air Traffic Management Mode C Altitude Reporting Mode of Secondary Radar (ICAO) Mode S SSR selective mode of interrogation NATS National Air Traffic Services (UK) OPS Operations PAN-OPS Procedures for Air Navigation - Operations (ICAO) RA Resolution Advisory RF Radio Frequency RWY Runway SARPS Standard And Recommended Practices (ICAO) SISG Safety Improvement Sub-Group SSR Secondary Surveillance Radar TCAS Traffic Collision Avoidance System

EUROCONTROL 46 EVAIR Summer Seasons 2006-2009 Evolution

EVAIR_bulletin04.indd 46 15/02/10 14:15 CONTACT US

The contact person for the EVAIR team is:

Ms Dragica Stankovic ATM Safety Expert EVAIR Function Manager [email protected] Tel: +32 2 729 5034 Fax: +32 2 729 9082

EVAIR Summer Seasons 2006-2009 Evolution 47 EUROCONTROL

EVAIR_bulletin04.indd 47 15/02/10 14:15 EUROCONTROL

February 2010 - © European Organisation for the Safety of Air Navigation (EUROCONTROL)

This document is published by EUROCONTROL for information purposes. It may be copied in whole or in part, provided that EUROCONTROL is mentioned as the source and it is not used for commercial purposes (i.e. for financial gain). The information in this document may not be modified without prior written permission from EUROCONTROL.

www.eurocontrol.int

EVAIR_bulletin04.indd 48 15/02/10 14:15