Review of UK Confidential Human Factors Incident Reporting
Programme (CHIRP) Data For The JAA/FAA Rotorcraft Human Factors
Study Group (RHFSG).
1. Introduction
CHIRP is a charitable trust located at Farnborough in the UK that provides a means for pilots, engineers, air traffic controllers and cabin attendants to report human factors related incidents in confidence. Details of the scheme are given in Annex 1 to this report.
The reports contained in the data base cannot always be fully validated and there is no practical way of establishing the extent to which individual reports have been validated. Data from this source must therefore be treated as anecdotal evidence. In addition, there is no practical means of associating the reports with flight hours or sectors so no rate information can be derived.
Objective data on the human factors aspects of accidents and incidents is rarely available, however, so most information in this field is of an anecdotal nature. The CHIRP data is therefore considered to be valuable and helpful despite its limitations.
2. Top Level Analysis
The CHIRP data base was interrogated on 06 November 2003 in support of the RHFSG exercise. The total number of reports on the system at that time was 4018, which were distributed amongst the four categories of air transport (AIR), engineering (ENG), general aviation (GA) and cabin issues (CABIN) as follows:
AIR (operational from 01 December 1982) = 3059
ENG (operational from 17 June 1997) = 315
GA (operational from 13 November 1996) = 425
CABIN (operational from 05 February 1998) = 219
Since no more detailed categorisation of the data is performed by CHIRP, further breakdown has to be performed by conducting word searches. All four categories were filtered using the word “helicopter” and the following numbers of reports identified:
AIR HELICOPTER = 200
ENG HELICOPTER = 7
GA HELICOPTER = 37
CABIN HELICOPTER = 2
In view of its size, the Air section of the data base was also filtered using the words “North Sea”, “rotor” and “helideck” and the following numbers of reports identified:
AIR NORTH SEA = 79
AIR ROTOR = 51
AIR HELIDECK = 21
It is quite possible that further word searches may have identified additional, relevant reports. The numbers likely to be involved are, however, considered to be insignificant.
3. Detailed Analysis
3.1 General
All 397 reports identified were individually reviewed, and duplicates eliminated along with reports covering matters not relevant to the study. Reports of a general nature not directly related to an incident were also removed. The remaining 72 reports were then categorised according to whether they addressed maintenance or design issues, and allocated to a category in the RHFSG scheme of 28 April 2004 (see Annex 2). Although not always explicitly stated, it was possible to confidently assign all but one of the reports to the small (<3175kg) and large (>3175kg) helicopter weight classes. The ‘doubtful’ report (no. 2038) was assigned to the large helicopter class, giving a total of 15 small helicopter reports and 57 large.
In common with incident reports where the information available is limited and the full context sometimes unknown, it can be difficult to be certain of the precise nature of the problem. In addition, trying to fit the reports to a fixed classification scheme will often result in compromises; sometimes a suitable category does not exist, and at other times the data appears to cover more than one category. For the purposes of the RHFSG exercise, the reports were allocated to the issues established in the RHFSG scheme wherever it was considered reasonable. Where more than one issue was covered, the report was allocated to what was considered to be the primary issue, and any secondary issue(s) not otherwise highlighted were noted in Annex 2 and Annex 3 for information.
The 72 reports reviewed are reproduced in Annex 3 to this report. Each entry comprises the CHIRP reference number, the section of the data base and keyword used to filter the data, the classifications allocated, and the verbatim text of the report as it appears on the CHIRP data base.
3.2 Maintenance and design issues
Of the 72 reports reviewed, 44 were considered to be fully relevant and 28 potentially relevant. These were divided into maintenance and design issues by weight class as indicated in the following tables:
Small helicopters (<3175kg):
Relevant(ü) / Potentially relevant (?) / Irrelevant
(x) / Total
Maintenance / 2 / 1 / 12 / 15
Design / 8 / 6 / 1 / 15
Large helicopters (>3175kg):
Relevant(ü) / Potentially relevant (?) / Irrelevant
(x) / Total
Maintenance / 2 / 0 / 55 / 57
Design / 35 / 22 / 0 / 57
Design issues clearly dominate the CHIRP reports for both small and large helicopters and the reasons for this are revealed in Section 3.3 below.
3.3 RHFSG classification
3.3.1 General:
The RHFSG classification system comprises four sections:
1. Operational issues identified from fixed-wing experience.
2. Issues identified from fixed-wing experience (NPA 25-310).
3. Known rotorcraft specific HF issues.
4. Other issues arising from data analysis.
Each section contains a number of issues against which the reports are compared. An entry is then made against each issue in each section to which one or more reports were allocated according to the following scheme:
Class 1 – Issue identified as a causal factor in a fatal accident.
Class 2 – Over 10% of HF issues attributed to this cause.
Class 3 – Other HF causal factors.
NB: Possibly because the general aviation section of CHIRP has been running for only about one third the time of the air transport section, the number of reports relating to small helicopters are quite low. In order to avoid undue emphasis being accorded to individual small helicopter reports, the percentages for both the small and large helicopter categories have been calculated using the total number of relevant reports (72).
3.3.2 Class 1 issues
By definition, CHIRP reports would not normally be expected to relate to fatal accidents since these will usually be comprehensively covered by a full accident report. Two reports (1113 and 1164) however, do relate to the same problem that had been the cause of a fatal accident and the corresponding issue (1.14 – misuse of adjacent controls) is therefore set to Class 1 for small helicopters.
3.3.3 Class 2 issues:
In the context of this report, Class 2 has been allocated to the three issues covered by 10% or more of the 72 CHIRP reports reviewed. All three issues qualify for the Class 2 rating for large helicopters only. The three issues are:
3.2 Flight deck environment – 19 reports (26.4%)
2.4 Human error with no associated technical failure – 9 reports (12.5%)
3.1 Low visibility rig approaches – 8 reports (11.1%)
Flight deck environment issues (3.2) are clearly the most prominent in the CHIRP database by a significant margin, and are predominantly related to offshore commercial helicopter operations. Review of the corresponding reports reveals sub-issues of: immersion suits; lack of air conditioning; noise; vibration; lack of sun visors; lack of heaters; uncomfortable seats; poor air quality. Whereas these issues may individually appear relatively innocuous, when coupled with long duty periods they can easily lead to excessive fatigue and hence human error. There are also reports of one crew member waking up to find the other asleep. One solution might be to improve the flight deck environment; another approach may be to reduce the maximum allowable duty time.
The reports allocated to the “human error with no technical failure” category cover a range of problems. In six of the nine reports (253, 526, 552, 563, 625, 968), however, the issues of flight crew fatigue and crew duty time are mentioned. Since there is good evidence to suggest that these issues are linked to flight deck environment, there is a case for including these six reports in the flight deck environment category, further increasing its importance.
The relatively high incidence of reports relating to low visibility rig approaches or airborne radar approaches (ARAs) suggests this to be an area of particular concern to pilots involved in offshore commercial helicopter operations. Six of the eight reports on the subject were associated with incidents; two related to inadvertent loss of height (342 & 2207), two related to near misses with the platform (496 & 610), one concerned loss of sight of the platform and a late go-around (675), and one concerned the loss of the platform target on the radar display (631). Although this is an operational issue, it should be borne in mind that the only practical solution lies in the development of appropriate airworthiness requirements. A robust technical solution to the conduct of low visibility rig approaches is required. The UK CAA has already recognised that the aircraft’s weather radar is neither designed nor certificated for the purpose for which it is being used in this instance.
3.3.4 Class 3 issues:
Class 3 has been allocated to the remaining issues covered by the CHIRP reports as indicated in Annex 2.
Review of the CHIRP reports prior to the establishment of the RHFSG classification scheme had identified human-machine interface (HMI) issues as a significant concern. Fifteen (20.8%) reports were considered relevant and one (1.4%) potentially relevant as shown in the following table.
RHFSG classification / RHFSG class / CHIRP report numbers / No. of reports / Total number of reports in classification / % of classification related to HMI issues<3175kg / >3175kg
1.6 / 3 / 240 (?) / 1 / 1 / 100%
1.7 / 3 / 1581 / 684, 953 / 3 / 3 / 100%
1.14 / 1 / 1113, 1164 / 2 / 2 / 100%
1.15 / 3 / 1067 / 799, 801 / 3 / 3 / 100%
2.2 / 3 / 537 / 1 / 1 / 100%
2.4 / 2 / 625, 800 / 2 / 12 / 17%
3.17 / 3 / 720, 1490 / 2 / 2 / 100%
3.19 / 3 / 1594, 2250 / 2 / 3 / 67%
It is noted that the HMI related reports account for 100% of the reports in six of the eight RHFSG classifications concerned, and most of the reports in a seventh. Effectively spreading the HMI issues across a number of (mostly class 3) classifications appears to have diluted the importance of this issue, and it is recommended that it be considered a high priority issue. It should also be noted that the associated CHIRP reports are quite evenly shared between the small and large helicopter categories, suggesting that this issue is not confined to offshore commercial helicopter operations.
3.3.5 Other Issues:
Four of the 72 reports could not sensibly be allocated to any of the RHFSG categories. Of these, one related to duty hours (480), one to the burden of regulation (1532), one to check list deficiencies (1935), and one to pilot training matters (2106).
4. Conclusions
In the context of the RHFSG exercise, the CHIRP data is considered to represent a valid input as anecdotal evidence.
In terms of the RHFSG criteria, the following four issues were ranked as either Class 1 or Class 2, and therefore represent areas of major concern:
1.14 - Misuse of adjacent controls.
3.2 - Flight deck environment.
2.4 - Human error with no associated technical failure.
3.1 - Low visibility rig approaches.
Two thirds of the reports allocated to issue 2.4 are considered to be related to flight deck environment through flight crew fatigue.
In addition, HMI issues (potentially including RHFSG classification issues 1.6, 1.7, 1.14, 1.15, 2.2, 2.4, 3.17 and 3.19) collectively accounted for 22.2% of the CHIRP reports and should therefore also be considered a major area of concern.
D A Howson
UK CAA
Research Management Dept
Annex 1
The UK Confidential Human Factors Incident Reporting Programme
Introduction
Although accident/incident rates in commercial air transport operations have reduced to an extremely low level, the number of accidents with Human Factors causes has not declined at the same rate and thus has become the dominant cause in major accidents.
Incident reporting programmes have proved to be valuable tools in the identification of safety related issues and the definition of corrective actions. In the specific case of incidents involving human error, the availability of an independent, voluntary, confidential reporting medium has provided valuable additional information to that available through formal or mandatory reporting systems.
The CHIRP Charitable Trust
The UK Confidential Human Factors Incident Reporting Programme, more commonly known by the acronym CHIRP, has been operating since 1982. In 1996 following a comprehensive independent review by the Guild of Air Pilots and Air Navigators (GAPAN), the Programme was restructured to enable it to make a more effective contribution to the resolution of important safety related issues in the UK air transport industry.
CHIRP was established in its present form, as a charitable company limited by guarantee, on 1 November 1996. This corporate structure was selected in order to provide a totally independent organisation, with management and fiscal responsibilities held by an independent Board of Trustees. The Programme is funded by the Civil Aviation Authority, which in turn is funded by the air transport industry.
The current Board of Trustees comprises 12 independent members and the Chief Medical Officer of the Civil Aviation Authority, who is appointed on an ex-officio basis. The Chairman is Professor Tony Nicholson, an acknowledged specialist in Aviation Medicine, and formerly Commandant of the Royal Air Force School of Aviation Medicine. The remaining Trustees are: