0
European Helicopter Safety Team - EHEST EHEST Analysis of 2006-2010 European Helicopter Accidents Final EHSAT Analysis Report Report Document ref. Status Date Final 2015-08-11 Contact details for enquiries: European Helicopter Safety Team ehest@easa.europa.eu www.easa.europa.eu/essi Disclaimer: The safety improvement analyses and recommendations produced by EHEST are based on expert judgement and are supplementary to the official reports of the accident investigation boards (AIB). Such recommendations, and the safety improvement actions that may follow, are solely aimed at improving helicopter safety, are not binding and under no circumstances should be considered to take precedence over the official AIB reports. The adoption of such safety improvement recommendations is subject to voluntary commitment, and engages only the responsibility of those who endorse these actions. The EHEST accepts no responsibility or liability whatsoever with regard to the content or for any actions resulting from the use of the information contained in these recommendations. Name Coordinated by Reference group Reviewed by Martin Bernandersson (EASA) Tony Eagles (UK CAA), Marc Greiller (Airbus Helicopters), Michel Masson (EASA), John Steel (IAA) Andy Evans (Aeroassurance), John Franklin (EASA), Yngvi Rafn Yngvason (EASA) 1
Table of Contents Executive Summary... 3 1 Analysis Results... 4 1.1 Basic Data... 4 1.2 SPS Analysis Time Period Comparison... 8 1.3 SPS Analysis Type of Operation Comparison... 10 1.4 SPS Level 2 Analysis Commercial Air Transport... 12 1.5 SPS Level 2 Analysis Aerial Work... 16 1.6 SPS Level 2 Analysis General Aviation... 18 1.7 HFACS Level 2 Analysis Commercial Air Transport... 20 1.8 Intervention Recommendations Analysis... 22 2 Concluding Remarks and Way Forward... 26 2
Executive Summary This report covers the work performed by the European Helicopter Safety Analysis Team (EHSAT), a sub-group of the European Helicopter Safety Team (EHEST) the rotorcraft pillar of the European Strategic Safety Initiative (ESSI) and the European component of the International Helicopter Safety Team (IHST). This report is a follow-up on the first EHSAT report published in 2010 addressing 2000-2005 European Helicopter accidents. It covers the analyses performed by the EHSAT regional teams of accidents that occurred in the years 2006-2010. It also provides comparisons between the two time periods 2000-2005 and 2006-2010 as well as some deeper analysis of results covering the entire time period 2000-2010. EHSAT Regional Teams were established in the following countries: Finland, France, Germany, Hungary, Ireland, Italy, the Netherlands, Norway, Spain, Sweden and the United Kingdom. The analysis methodology used is the Standard Problem Statements (SPS) and Human Factors Analysis and Classification System (HFACS). This methodology was presented in detail in the first EHSAT report. For further information please consult that publication. The continued analysis of Helicopter Accidents in the period 2006-2010 by the EHSAT Regional Teams and the comparison with the analysis of 200-2005 accidents have generally confirmed that the issues identified in this first period continue to be of concern and that the safety improvement actions decided and developed based on the first analysis period were still valid. A few differences have been identified though and additional analyses have been conducted. These will help shaping up the future priorities of the European Helicopter Safety Implementation Team (EHSIT) and its three Specialist Teams on Training, Ops & SMS and Technology. This report will also be communicated to EASA and contribute developing the helicopter Safety Risk Portfolio (SRP), which will help populating the helicopter section of the European Aviation Safety plan (EASp). Additional data on helicopter safety in Europe can also be found in the EASA Annual Safety Reviews published annually on the EASA website. 3
1 Analysis Results 1.1 Basic Data Figure 1 shows the number of accidents that were analysed by the EHSAT Teams, broken down by year of occurrence. 70 Number of analysed accidents per year of occurrence 60 Number of analysed accidents 50 40 30 20 10 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Year Figure 1 Number of analysed accidents per year of occurrence In the period 2000-2005, 325 accidents were analysed. Some national teams were unable to continue completing the analysis of accidents for the period 2006-2010, and therefore only 162 accidents were analysed in that period. In the period 2006-2010, 537 helicopter accidents occurred in the EASA Member States (Source: EASA ADREP Database), which means that the EHSAT teams analysed 30% of all accidents that occurred in that time period. As not all of the 537 accidents have been investigated by an Accident Investigation Board, it would therefore not have been possible to analyse all those accidents using the EHSAT methodology. However, the most serious accidents in the participating states are likely to have been investigated by AIBs and analysed by the EHSAT Regional teams. Figure 2 shows the distribution of top level Operation Types amongst the 487 analysed accidents. 4
Operation type distribution 2000 2010 accidents AW CAT GA State 4% 30% 44% 22% Figure 2 Operation Type distribution 20% of the analysed accidents were Commercial Air Transport operations, 31% were Aerial Work, 45% were General Aviation and 4% were non-military State Flights. As fleet usage data per type of operation on a European Level is not available, it has not been possible to assess whether any type of operation has a differing share of accidents compared to, for example, number of take-offs. Figure 3 shows the distribution of Highest Injury Level recorded for the 487 analysed accidents. Highest Injury Level distribution, 2000 2010 accidents Fatal Serious Minor None Unknown 0% 24% 43% 13% 20% Figure 3 Highest Injury Level distribution 5
24% of the analysed accidents were fatal, i.e. one or more persons involved in the accident died as a result of the injuries sustained in the accident. On the other hand, in 43% of the accidents there were no injuries. In 13% of the accidents the persons involved sustained serious injuries, and in 20% of the accidents minor injuries were sustained. Figure 4 shows the distribution of aircraft damage levels in the 487 analysed accidents. Aircraft Damage distribution 2000 2010 accidents Destroyed Substantial Minor None 3% 5% 48% 44% Figure 4 Aircraft damage distribution Damage levels are derived from the accident reports. It is worth noting that in 46% of the accidents where the aircraft was destroyed, one or more persons involved sustained fatal injuries. Figure 5 shows the distribution of the phase of flight in which the accident occurred. 6
Phase of Flight distribution, 2000 2010 accidents Standing Taxi Take off En route Hover Manoeuvring Approach Landing 18% 4% 2% 15% 10% 23% 1% 27% Figure 5 Phase of Flight distribution The Manoeuvring phase involves intentional low level, low speed flying in the vicinity of obstacles and is the phase where most of the Aerial Work accidents occurred (39%). Also noteworthy is that 61% of all fatal accidents occurred in the En-route phase, whilst most accidents with serious injuries occurred in the Manoeuvring phase (25%). Figure 6 shows the distribution of engine configurations of the analysed accidents. Engine Configuration distribution 2000 2010 accidents Single Piston Multi Piston Single Turb. Multi Turb. 16% 40% 43% 1% Figure 6 Engine configuration distribution 7
Most analysed accidents occurred to single engine helicopters, who also comprise the majority of the helicopter fleet in Europe. 1.2 SPS Analysis Time Period Comparison In this section, comparisons will be made between the initial time period of 2000-2005 and the second period of 2006-2010 to see whether there are any significant differences between the two time periods. In order to facilitate easy comparison and reduce the impact of differing implementation of the coding instructions, the percentage of accidents where the SPS/HFACS codes on Level 1 have been assigned at least once will be used in this section, as was the case in the previous report. Figure 7 shows the percentage of analysed accidents where SPS codes on the top level (Level 1) was assigned at least once, comparing the periods 2000-2005 with 2006-2010. Percentage of analysed accidents where SPS Level 1 was assigned at least once SPS Level 1 Pilot judgment & actions Safety Management Ground Duties Pilot situation awareness Data issues Mission Risk Part/system failure Regulatory Aircraft Design Post crash survival Communications Maintenance Infrastructure Ground personnel 2006 2010 2000 2005 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% Percentage Figure 7 Percentage of analysed accidents where SPS Level 1 was assigned at least once Figure 7 shows that there is a close correlation between the two analysed time periods; the same areas are still of concern and the issues that were identified in the analysis of the 2000-2005 accidents continue to be significant and the proposed mitigations are still valid. The highest level of Standard Problem Statements, level 1, only provides information on a general level. To better understand what kind of factors played a role in the accident data set one must look at a deeper level in the taxonomy, which will be done later in this report. Figure 8 shows the distribution of the percentage of analysed accidents where the top level HFACS codes have been assigned at least once. 8
Percentage 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% Percentage of analysed accidents where HFACS Level 1 was assigned at least once 2000 2005 2006 2010 HFACS Level 1 Figure 8 - Percentage of analysed accidents where HFACS Level 1 was assigned at least once Also in Figure 8 the magnitude of the HFACS Level 1 codes correlate between the two periods. This is an indication that the problem areas and proposed mitigations identified in the first time period are still valid. Figure 9 shows the percentage of analysed accidents where HFACS Maintenance Extension (ME) Level 1 codes were assigned at least once. 9
Percentage of analysed accidents where HFACS ME Level 1 was assigned at least once 9.0% 8.0% 7.0% Percentage 6.0% 5.0% 4.0% 3.0% 2.0% 2000 2005 2006 2010 1.0% 0.0% HFACS ME Maintainer Acts HFACS ME Maintainer Conditions HFACS ME Working Conditions HFACS ME Management Conditions HFACS ME Level 1 Figure 9 - Percentage of analysed accidents where HFACS ME Level 1 was assigned at least once It should be noted that the number of accidents where Human Factors issues in maintenance were identified is small, slightly less than 10% of the analysed accidents had an HFACS ME code assigned. Also noteworthy is the fact that accident investigations seldom go into the depth of identifying and analysing Human Factor issues in maintenance operations. 1.3 SPS Analysis Type of Operation Comparison Figure 10 shows the percentage of accidents where SPS Level 1 has been assigned at least once, split into types of operation (General Aviation, Aerial Work, Commercial Air Transport) for the whole time period 2000-2010. 10
Percentage of accidents with SPS Level 1 at least once 2000 2010 GA AW CAT SPS Level 1 Aircraft Design Regulatory Ground personnel Data issues Post crash survival Mission Risk Part/system failure Pilot situation awareness Communications Pilot judgment & actions Infrastructure Maintenance Safety Management Ground Duties 0% 10% 20% 30% 40% 50% 60% 70% 80% Percentage Figure 10 - Percentage of analysed accidents per type of operation where SPS Level 1 was assigned at least once Noteworthy is that Pilot Situation Awareness is featured most frequently in Commercial Air Transport accidents. Mission Risks shows the biggest difference between the three types of operation. The high presence of Mission Risk SPS in Aerial Work is normal considering that Aerial Work operations are often complex, higher risk missions. The high presence of Data Issues in General Aviation is also not surprising since Accident Investigations into General Aviation accidents normally don't go into as deep detail as for other accidents. Figure 11 shows the percentage of accidents where HFACS Level 1 codes were assigned at least once, split by types of operation. 11
Percentage of accidents with HFACS Level 1 at least once 2000 2010 GA AW CAT HFACS Organizational Influences HFACS Supervision HFACS Level 1 HFACS Preconditions Personnel Factors HFACS Preconditions Condition of Individuals HFACS Preconditions Environmental Factors HFACS Unsafe Acts Violations HFACS Unsafe Acts Errors 0% 10% 20% 30% 40% 50% 60% 70% Percentage Figure 11 - Percentage of analysed accidents per type of operation where HFACS Level 1 was assigned at least once The number of accidents where HFACS ME were identified is so small that it is deemed insignificant and is not further analysed in this report. 1.4 SPS Level 2 Analysis Commercial Air Transport The analysis in this and the following sections are based on total number of SPS assigned and is split by type of operation. A selection of SPS on Level 1 have been analysed deeper into Level 2 to highlight areas of concern. The Level 2 distribution for the Level 1 SPS "Pilot Judgement & Actions" is shown in Figure 12. 12
CAT SPS "Pilot Judgement & Actions" SPS Level 2 Distribution Crew Resource Management 7% Landing Procedures 12% Procedure Implementation 12% Human Factors Pilot's Decision 43% Flight Profile 17% Human Factors Pilot/Aircraft Interface 9% Figure 12 CAT SPS "Pilot Judgement & Actions" SPS Level 2 Distribution Pilot decision making is the largest concern in this area. It should be noted that the relatively small share of Crew Resource Management (CRM) issues is probably influenced by the majority of flights being flown in a single-pilot environment, approximately 70% of the analysed CAT accidents were flown in a single-pilot environment. The Level 2 distribution of the SPS Level 1 "Safety Management" is shown in Figure 13. CAT SPS "Safety Management" SPS Level 2 Distribution Ground/Passenger Training 0% Survival training 1% Inadequate Pilot Experience 20% Transition Training 1% Flight Procedure Training 9% Management 31% Scheduling/Dispatch 1% Training Program Management 5% Pilot 14% Equipment (Safety Management) 4% Safety Program 14% Figure 13 CAT SPS "Safety Management" SPS Level 2 Distribution 13
The Management of the operations and the assignment of inexperienced pilots to difficult missions are the most prominent concerns in this area. The Level 2 distribution of the SPS Level 1 "Pilot Situation awareness" is shown in Figure 14. CAT SPS "Pilot Situation Awareness" SPS Level 2 Distribution Visibility/Weather 43% External Environment Awareness 50% Internal Aircraft Awareness 7% Figure 14 CAT SPS "Pilot Situation Awareness" SPS Level 2 Distribution The two most prominent concerns in this area are the External Environment Awareness and the Weather and Visibility. The Level 2 distribution of the Level 1 SPS "Ground Duties" is shown in Figure 15. 14
CAT SPS "Ground Duties" SPS Level 2 Distribution Weight and Balance Preflight Briefings 4% 9% Aircraft Preflight 7% Mission Planning 80% Figure 15 CAT SPS "Ground Duties" SPS Level 2 Distribution Mission Planning is the area in which most Ground Duties SPS have been identified. The Level 2 distribution of the SPS Level 1 "Part/System failure" is shown in Figure 16. CAT SPS "Part/System Failure" SPS Level 2 Distribution Part/system failure Operational FOD Part/system failure 3% Powerplant 11% Part/system failure Mission Specific Equipment 3% Part/system failure Aircraft 83% Figure 16 CAT SPS "Part/System Failure" SPS Level 2 Distribution 15
Failures of the airframe (rather than powerplant/engine failures) are the highest number identified in this area. 1.5 SPS Level 2 Analysis Aerial Work This section deals with the SPS on Level 2 for Aerial Work, for selected Level 1 SPS. Figure 17 shows the Level 2 distribution of the Level 1 SPS "Pilot Judgement and Actions". AW SPS "Pilot Judgement & Actions" SPS Level 2 Distribution Crew Resource Management 3% Procedure Implementation 10% Human Factors Pilot's Decision 27% Landing Procedures 16% Flight Profile 23% Human Factors Pilot/Aircraft Interface 21% Figure 17 AW SPS "Pilot Judgement & Actions" SPS Level 2 Distribution The top 3 issues identified in this area are Pilot's Decision, Flight Profile and Pilot/Aircraft Interface, with the latter noticeably higher than for CAT and GA. Figure 18 shows the Level 2 distribution of the Level 1 SPS "Safety Management" for Aerial Work. 16
Scheduling/Dispatch 4% AW SPS "Safety Management" SPS Level 2 Distribution Ground/Passenger Training Inadequate Pilot 0% Experience 14% Transition Training 1% Flight Procedure Training 9% Training Program Management 3% Pilot 9% Management 31% Equipment (Safety Management) 16% Survival training 3% Figure 18 AW SPS "Safety Management" SPS Level 2 Distribution In this area, the top 3 identified issues are Management, Equipment (Helicopters not adequately equipped for mission purposes and/or Personal protective equipment inadequate or unavailable) and Safety Program. Management is overall the third highest SPS Level 2 for Aerial Work. Figure 19 shows the Level 2 distribution of the Level 1 SPS "Mission Risk". Safety Program 10% AW SPS "Mission Risk" SPS Level 2 Distribution Aircraft Intensive 3% Environment 6% Crew Intensive ( e.g. winching, HEMS, load lifting etc.) 0% Pilot Intensive 33% Terrain/Obstacles 58% Figure 19 AW SPS "Mission Risk" SPS Level 2 Distribution 17
Not surprisingly, the proximity to terrain and obstacles are the highest areas of concerns for Aerial Work missions. Many Aerial Work missions are conducted at low height, for example Power line inspections, Reindeer herding and firefighting. 1.6 SPS Level 2 Analysis General Aviation This section deals with the SPS on Level 2 for General Aviation, for selected Level 1 SPS. Figure 20 shows the Level 2 distribution of the Level 1 SPS "Pilot Judgement & Actions". GA SPS "Pilot Judgement & Actions" SPS Level 2 Distribution Procedure Implementation 16% Crew Resource Management 6% Human Factors Pilot's Decision 41% Landing Procedures 13% Flight Profile 16% Human Factors Pilot/Aircraft Interface 8% Figure 20 GA SPS "Pilot Judgement & Actions" SPS Level 2 Distribution In this area, the top 3 issues are Pilot's decision, Flight Profile and Procedure Implementation. Figure 21 shows the Level 2 distribution of the Level 1 SPS "Safety Management". 18
Inadequate Pilot Experience 36% GA SPS "Safety Management" SPS Level 2 Distribution Ground/Passenger Training 0% Transition Training 3% Flight Procedure Training 17% Survival training 1% Management 12% Safety Program 5% Equipment (Safety Management) 3% Figure 21 GA SPS "Safety Management" SPS Level 2 Distribution In this area, the top 3 issues are Inadequate Pilot Experience, Flight Procedure Training and Pilot (disregard of known safety risk/self-induced pressure), with, as to be expected, less management issues in GA. Pilot 12% Scheduling/Dispatch 1% Training Program Management 10% Figure 22 shows the Level 2 distribution of the Level 1 SPS "Ground Duties". GA SPS "Ground duties" SPS Level 2 Distribution Aircraft Preflight 10% Preflight Briefings 7% Postflight Duties 1% Weight and Balance 11% Mission Planning 71% Figure 22 GA SPS "Ground Duties" SPS Level 2 Distribution 19
Mission Planning is the main issue in this area, followed by Weight and Balance and Aircraft preflight. 1.7 HFACS Level 2 Analysis Commercial Air Transport This section highlights the HFACS analysis performed by the regional teams on Commercial Air Transport accidents. Figure 23 shows the Level 2 distribution of the Level 1 HFACS "Preconditions Conditions of Individuals". CAT HFACS "Preconditions Conditions of individuals" HFACS Level 2 Distribution Adverse Physiological States Physical/Mental 8% Limitations 9% Cognitive Factors 38% Perceptual Factors 17% Psycho Behavioural Factors 28% Figure 23 CAT HFACS "Preconditions Conditions of individuals" HFACS Level 2 Distribution Cognitive Factors, Psycho-Behavioural Factors and Perceptual Factors are the 3 main issues in this area. Figure 24 shows the Level 2 distribution of the Level 1 HFACS "Unsafe acts Errors". 20
CAT HFACS "Unsafe acts Errors" HFACS Level 2 Distribution Perceptual Errors 11% Skill based Errors 40% Judgement & Decision Making Errors 49% Figure 24 CAT HFACS "Unsafe acts Errors" HFACS Level 2 Distribution The Judgement and Decision-making Errors are the main issues in this area, followed by the Skill-Based Errors. Figure 25 shows the Level 2 distribution of the Level 1 HFACS "Organisational Influences". CAT HFACS "Organisational Influences" Level 2 Distribution Resource/Acquisition Management 27% Organizational Process 56% Organizational Climate 17% Figure 25 CAT HFACS "Organisational influences" Level 2 Distribution 21
The Organisational Process, which includes Workload issues, is the top issue of concern in this area. Figure 26 shows the Level 3 distribution of the Level 1 and Level 2 HFACS "Violations". There is no Level 2 HFACS categorisation under "Violations". CAT HFACS "Violations" Level 3 Distribution Violation Lack of Discipline 29% Violation Based on Risk Assessment 28% Violation Routine/Widespread 43% Figure 26 CAT HFACS "Violations" Level 3 Distribution 1.8 Intervention Recommendations Analysis This section reviews the Intervention Recommendations (IRs) assigned by the EHSAT national teams. All IRs that have been assigned in the analysis of the 2006-2010 accidents have been collected and sent to the respective EHSIT implementation teams for further processing. This section covers IRs assigned to accidents in the time period 2000-2010. IRs are categorised in two levels, and Figure 27 shows the number of IRs per Level 1 category. 22
Number of IRs Level 1 All accidents 2000 2010 800 700 600 500 400 300 200 100 0 Figure 27 Number of IRs Level 1 All accidents 2000-2010 The top 3 categories of Intervention Recommendations are Operations, Training/Instructional and Regulatory. This was already seen in the analysis of the 2000-2005 accidents and the EHSIT teams were formed accordingly. Figure 28 shows a time period comparison on the average number of IRs assigned per accident. The average has decreased slightly in the 2006-2010 period. 6 Average number of IRs per accident Number of IRs per accident 5 4 3 2 1 0 2000 2005 2006 2010 Year Figure 28 Time period comparison Average number of IRs per accident The Intervention Recommendations are scored, based on the judgement of the EHSAT Analysis Team performing the analysis, on their Ability and Usage. Ability being how effective the proposed intervention would be in mitigating the problem, and Usage being the how probable it is that the proposed intervention will be implemented. Both Ability and Usage are 23
scored from 0 to 4. Figure 29 shows the percentage distribution of the Ability scoring for each IR Level 1 category. IR Ability scoring All accidents 2000 2010 0 1 2 3 4 Aircraft_Design Aircraft_System_Equipment_Design Data_or_Information_Issues Infrastructure Maintenance Manufacturing Operations Regulatory Research Search_and_Rescue Training_Instructional 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Figure 29 Distribution of IR Ability scoring per IR Level 1 category All accidents 2000-2010 The highest Ability IRs have been given in the Maintenance category, with more than 80% of those IRs scoring 3 or 4. Data or Information issues, which mainly covers the availability of data to accident investigators, also have more than 80% scoring 3 or 4. The lowest Ability IR category is the Search and Rescue, with no IRs scoring 3 or 4. It should however be noted that only 3 Search and Rescue IRs were proposed. Figure 30 shows the precentage distribution of the Usage scoring fore each IR Level 1 category. 24
IR Usage scoring All accidents 2000 2010 0 1 2 3 4 Aircraft_Design Aircraft_System_Equipment_Design Data_or_Information_Issues Infrastructure Maintenance Manufacturing Operations Regulatory Research Search_and_Rescue Training_Instructional 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Figure 30 Distribution of IR Usage scoring per IR Level 1 category All accidents 2000-2010 The Usage scoring is generally lower than the Ability scoring. Infrastructure IRs (Search and Rescue excluded) have the lowest usage scoring, with only just over 30% scoring 3 or 4. Many IR categories, for example Operations, Training/Instructional and Data/Information Issues also have just over 30% scoring 3 or 4. The highest usage scorings are in the Manufacturing and Regulatory categories with over 50% scoring 3 or 4. 25
2 Concluding Remarks and Way Forward The continued analysis of Helicopter Accidents in the period 2006-2010 by the EHSAT Teams and the comparison with the analysis of 2000-2005 accidents have confirmed that the issues identified in this first period continue to be of concern and that the safety improvement actions decided based on the first analysis period are still valid. Since this first analysis report was published, the EHSIT Specialist Teams Training, Ops & SMS and Technology have produced a number of safety promotion deliverables in the form of leaflets, videos, toolkits, manuals and reports. That material addresses and provides ways to mitigate the top safety issues and intervention recommendations identified in the analysis of the 2000-2005 accidents. This work continues and the results of the analysis of the 2006-2010 accidents will be communicated to the EHSIT STs and will contribute to shaping the future priorities and actions of the EHIST Specialist Teams. Results will also be shared within EASA and contribute to defining the helicopter Safety Risk Portfolio, which will serve as a basis to develop the helicopter section of the European Aviation Safety plan (EASp). 26