Desktop version

Home arrow Management

  • Increase font
  • Decrease font


<<   CONTENTS   >>

Why Historical Models Are Problematic

Table of Contents:

The models we have looked at represent a challenge for developing a competence- based approach to CRM. I want to illustrate the problem by following one airline’s experience of coping with operational performance issues. The starting point for the discussion is a tragic, fatal accident. On 12 January 1999, an F27 cargo aircraft carrying newspapers crashed after it stalled on the approach to Guernsey Airport, in the Channel Islands (AAIB. 2000). The immediate cause of the crash was the load shifting in flight. As a result, when the crew selected the landing flap, the aircraft’s centre of gravity moved out of limits, and the aircraft stalled. The aircraft had been, first, incorrectly loaded and then, second, the load had not been properly secured, which is why it shifted in flight.

To fully understand the actual ‘cause’ of the accident, we need to go back a few years. The company originally operated between fixed bases with aircraft loading and administration being undertaken by company staff. The company then went through an expansion and took on a significant number of parcel contracts for the UK postal service. Parcels tend to fill the space available before they become weight limiting. Therefore, loading the aircraft was simple: you just packed in the parcels and threw a net over them. Weight and balance was never an issue. The next phase of growth saw the company moving into the ad hoc charter market. Cargo was collected from wherever it was available and ground handling was often undertaken by third parties who might never have dealt witli an F27 before. This was the situation on the day of the accident but, crucially, the company’s procedures for the oversight of loading aircraft had not kept up with the changes brought about by the growth and the opening up of new' markets. These changes in the nature of the loads carried and the ground support available to deal with loading meant that the airline’s risk profile had changed. In Chapter 1, we saw that an effective competence model should reflect the contingencies most likely to be encountered by the crew. In this case, commercial developments over time had changed the nature of the task and the company’s processes - and crew competence - had become uncoupled.

The response to the accident was remarkably lacking in insight. Because the aircraft stalled when the crew selected the landing flap, the procedures were changed to prohibit the future use of full flap for landing. Nothing was done about loading procedures. Of course, the landing flap is designed to reduce both the touchdown speed and the stopping distance needed on the runway. On the morning of 1 July 2000 (AAIB, 2001), just over a year later, a company F27 was on final approach to Coventry Airport. The weather was mist and light rain w'ith a cloud base of 400ft. With a slight taihvind, the aircraft was high on the glide slope and 19 kt fast. The captain touched down 600m along the runway with 1000 m remaining. The aircraft did not stop in the distance available. After that accident, the procedures were changed to restore full flap for landing. But the story did not end there. After the second accident, the company pilots were on their guard. One night, during an FO training flight, the captain said that they needed to be particularly careful about getting high and fast on the approach, his concern based on the folklore that had now developed in the airline. Neither of the pilots had been in the company w'hen the fatal crash occurred but were, of course, aw'are of the event. The result was a heavy landing almost in the undershoot as the crew ended up slow and low on the approach path (personal communication).

This sequence shows how the classical ‘barrier’ model results in constraints being modified or new barriers being put in place. The response to one event, ironically, created the context for the next. As well as formal, procedural, responses to the investigation process being introduced, a social change occurred based on the company’s expectations of tolerable crew' performance. Failure begat failure.

Conclusion

This chapter has looked at events from the perspective of accident investigation models that have been in common use for many years. While they all have undoubtedly served a purpose or provided some insight, neither strategy really addresses the problem of changing workplace behaviour to build robustness into processes. By default, almost, linear models seem to result in more barriers to be circumvented. By placing ever-more constraints on operations, efficiency becomes difficult to achieve unless short cuts are taken. The NAT and HRO schools are too abstract to be of practical use, and Reason’s model, too, lacks the specificity needed to generate inputs to training. In fairness, these various schools of thinking attempt to offer modes of explanation, not routes to change and, without a doubt, Reason’s approach was a milestone because it dragged the focus away from the individual to the broader organisational context. A competence-based approach to CRM will need different tools of analysis and models of performance and, in the next chapter, I want to outline an approach based on more recent thinking about safety.

References

AAIB. (2000). AAR 2/2000 Report in the Accident to Fokker F27-600 Friendship. G-CHNL near Guernsey Airport, Channel Islands, on 12 January 1999.

AAIB. (2001). AAIB Safety Bulletin AAR 4/2001 Channel Express, Coventry-Bagington Airport, UK.

Amalberti, R. (2001). Revisiting safety and human factors paradigms to meet the safety challenges of ultra complex and ultra safe systems’. In B. Wolpert & B. Fahlbruch (Eds.), Challenges and Pitfalls in Safety Interventions. Flolland: Elsevier, pp. 265-276

ATSB. (2001). Occurrence Report No: 199904317 dated April 10, 2001.

Blockley, D.I. (1992). Engineering Safety. London: McGraw-Hill.

Cox, S. & Cox, T. (1996). Safety, Systems and People. Oxford: Butterworth-Heinemann.

Cullen. W.D. (1990). The Public Inquiry into the Piper Alpha Disaster Vol 1 & 2. HMSO.

Dekker, S. (2002). The Field Guide to Human Error Investigations. Aldershot: Ashgate.

Hollnagel, E., Woods, D.D., & Leveson, N. (2006). Resilience engineering: Concepts and precepts. Aldershot, England: Ashgate.

Hollnagel. E. (2009). The ETTO Principle: Efficiency-thoroughness Trade-off. Why Things That Go Right Sometimes Go Wrong. Aldershot: Ashgate.

Leveson, N. (2011). Engineering a Safer World. Cambridge, MA: MIT Press.

MacLeod, N. (2005). Building Safe Systems in Aviation. Aldershot: Ashgate.

Marais, K., Dulac, N.. & Leveson, N. (2004). Beyond normal accidents and high reliability organisations: The need for an alternative approach to safety in complex systems. Paper presented at the Engineering Systems Division Symposium, March 29-31. Cambridge, M A: MIT.

NTSB. (1991). Report No: AAR-91/04 Avianca Boeing 707 Fuel Exhaustion Cove Neck, New York. January 25, 1990. Published April 30. 1991.

NTSB. (1993). Report No: FTW93-M-A143. In-flight Loss of Control. Continental Express Embraer 120. Pine Bluff. Arkansas. 29 April 1993.

Perrow, C. (1999). Normal Accidents. Princeton, NJ: Princeton University Press.

Reason, J. (1990). Human Error. Cambridge: Cambridge University Press.

Roberts, K.H. (1990). Some characteristics of high-reliability organisations. Organization Science, 1. 160-177.

Snook, S.A. (2000). Friendly Fire. Princeton, NJ: Princeton University Press.

SHK. (1993). Report C No: 1993:57. Air Traffic Accident on 27 December 1991 at Gottrora. AB County. Case 1-124/91.

TSB. (2005a). Aviation Investigation Report No: A0400336. Rejected Landing - Collision with Terrain. Oshawa Municipal Airport, Ontario, December 16, 2004. Released October 18. 2005.

TSB. (2005b). Aviation Investigation Report No: A04H0001. Loss of Control. Georgian Express Ltd. Cessna 208B Caravan C-FAGA Pelee Island, Ontario. 17 January 2004.

Tversky, A. & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124-1131.

 
<<   CONTENTS   >>

Related topics