International accident investigation is too open to accusations of partisanship. The industry must agree on a better system

The controversy sparked by the official report on the crash of Singapore Airlines (SIA) flight SQ006 at Taipei, Taiwan, has served to highlight, yet again, some of the fundamental problems with the established protocol for conducting air accident investigations. It is time the industry admits that the current system is inherently flawed and considers whether a better model could be devised.

Once again, the conclusions of a final report have polarised the "interested parties", each insisting its argument is based solely on the desire to advance aviation safety. Unfortunately, the travelling public is bound to conclude that politics have prevailed and - however painstaking the fact-finding may have been - the exercise is dominated by the motive to shift as much of the blame as possible into the other camp.

Taiwan's Aviation Safety Council (ASC) judged that, apart from the appalling weather conditions prevailing at the time, the SQ006 accident was caused purely by pilot error. The report also records - but not as causal factors - many shortcomings in the actions (or lack of them) of the airport authority.

Singapore's Ministry of Transport (MoT) denies that pilot error was the main cause, arguing the ASC had downplayed the fact that the airport had done nothing to make it impossible for pilots to mistake a partially closed runway for an active one.

In an extraordinary move which may become common practice in future air accident investigations unless the industry acts soon, the Singapore MoT has separately published its own report of the event, even though its comments were included fully in an annex to the ASC report. The annexing of additional comment is allowable procedure according to the International Civil Aviation Organisation, but it has become more common over the last five years. The annexed dissent, in the case of the recent Egyptair 990 and Silk Air accidents, constituted virtual alternative reports.

Singapore is also blasting the ASC for not allowing its accredited representatives to participate in the analysis phase of the investigation, despite the fact the ASC is adamant it acted fully in accordance with ICAO recommendations.

All this goes to show the need for a radical rethink of how the industry investigates accidents, derives safety lessons for the future (or doesn't), and then implements appropriate changes (or doesn't). It would be delightful to be able to be idealistic, but the sad truth is that the existing system, in reality, creates partisanship in all but the simplest cases.

The basic ICAO protocol is that the accident investigation is carried out, or at least headed by, agencies of the country where the event occurred. Unless that also happens to be the country in which the airline is based, the aircraft was designed and built, and of which the crew and passengers are nationals, partisanship can arise too easily.

The recently completed Egyptair flight 990 investigation was an example of how difficult it can be to reach a consensus when the reputations of aircraft manufacturers and national airlines could be construed as being at stake. In that instance Egypt refused to accept the US National Transportation Safety Board (NTSB) verdict that an Egyptair pilot deliberately caused the Boeing 767 crash, and appended its own theory based on mechanical failure.

Singapore was also involved in the inconclusive Indonesian probe into the loss of a Boeing 737 belonging to SIA subsidiary SilkAir in 1997. In that case, the US NTSB appended the conclusion that there was plenty of evidence the Singaporean pilot had deliberately crashed the aircraft.

There are two related dangers with the present system: the investigators might act in a partisan way and even if they do not they could too easily be accused of partisanship. No matter which happens, the likelihood of lessons being learned and applied is at least degraded and possibly destroyed.

It would be easy to trot out the argument that ICAO should carry out the investigation. In practice, however, it would be starved of resources. But it is feasible for ICAO, through its Assembly, to be voted active and legal charge over all international accident investigations. For that to happen it would need a cadre of accident investigators fully qualified to lead investigations, but the system for providing the resources and technical expertise could stay as it is. International sport has it right. Referees are always from a third country.

Source: Flight International