C.H. Simpson
Transportation Safety Board of Canada
ISASI, Victoria, B.C.
18 September 2001

Good Morning Ladies and Gentlemen.

It is indeed a personal privilege to be invited to address such a prestigious group as ISASI. Your Society has a professional membership dedicated to the single objective of advancing aviation safety. There is no doubt that the international nature of aviation is not only bringing us together at conferences like this, but consolidates our ideas and views on all aspects of the aviation industry.

Before continuing my remarks this morning I would like to acknowledge some of my colleagues from the TSB of Canada who are here with me this morning. Most of these people have been involved, under Vic Gerden's leadership, in the largest, most complex investigation ever undertaken by the TSB. Although the final report is far from complete, there has been considerable safety action, both informal and formal. The Board's interim recommendations to date have been well received by both the regulatory authorities and the aviation community. This has also been the most costly investigation in Canadian history - over 50 million dollars (the majority of it for search and recovery.) The financial resources are a Canadian responsibility, however, the value of the assistance from other countries was quite substantial, both during the immediate aftermath and during the continuing investigation. The human resources of the TSB were augmented in the safety investigation by an international team in an outstanding supportive role.

Major accidents increasingly rely on an international team effort, not only to investigate the accident, but to effect change throughout a global system that has few geographical or political boundaries.

An international team effort was both essential and highly visible during the investigation of SR111, and I feel that it is timely to talk about international cooperation. The TSB of Canada, has earned, I believe, a reputation in this area especially during our on-going investigation into the SR111 accident.

As per ICAO Annex 13, the State of Occurrence is primarily responsible for conducting the investigation. The fact that the problem is literally "in your backyard" and as such the national authority must deal with it, drives this agreement among States. However, as the international scope of accident investigation steadily increases, we continue to think more and more as an international community and find ways to globally improve safety as efficiently as possible. Many of you have considered ways and means to facilitate, coordinate and, ultimately, finance investigations into major accidents, and this is a step in the right direction. There is no question the financial liability of a major accident for a small country could be a serious deterrent to a full and successful investigation.

The reality of many of today's large major investigations is that the international community does, to a great extent, rise to the occasion. For SR111, the TSB had tremendous assistance, not only from other federal departments such as the RCMP, the Department of Defence, the Department of Fisheries and Oceans, the Canadian Coast Guard, provincial governments, local governments and many volunteers, but also from the United States NTSB, FAA and ALPA, the Swiss AAIB and CAA, and of course the operator and manufacturers involved . ICAO recognizes this "team response" and has done excellent work in promoting cooperation among States, and developing standards in an attempt to achieve some form of consistency between accident investigations, regardless of where they take place throughout the world.

However, despite these efforts, there is still a high degree of inconsistency, directly related to the State of Occurrence's resources and commitment to accident investigation on an on-going basis. There are only a handful of independent safety authorities in the world, with the majority of the organizations reporting to their national department of transport, and many organisations with a minimum number of trained investigators and often very limited capabilities. We all recognize that many States have difficulty justifying, and securing, the level of resources that the leading States have, for a variety of reasons; the reality remains that the same accident can have a significantly different level of effort depending on where the crash occurs.

For example ...

  • lack of financial resources necessary to investigate fully;
  • lack of human resources for the task;
  • the use of sensitive information such as cockpit voice recorder tapes (there is a major problem with the manner different States treat CVR recordings);
  • political ability to move recommendations beyond national borders (historically most recommendations from the State of Occurrence have limited effect beyond the national borders, unless supported fully by other States);
  • focus of an investigation varies;
  • the safety objective of an Annex 13 investigation can be mixed up with, and hindered by, criminal or civil litigation activity.

To some degree we are getting consistency in investigation approach in that, since Airbus is in France and Boeing is in the US, virtually all major accidents of large air carrier aircraft involve one of these two States. However, this is not by design and there would seem to be advantages to ensuring that at least one group is always involved in every major occurrence, as part of the investigation body. Investigations of major accidents are usually complex and expensive, and we must ensure they receive the amount of attention and resources needed to address aviation safety worldwide.

While we are talking about capabilities, there are only a few States that have flight recorder capabilities, and not just capability to readout the recorder (which is becoming very simple) but capability to effectively interpret and analyse voluminous data. It is no longer feasible to print the data out and pass it around to everyone, but rather sophisticated interactive software tools and a high level of experience are required to do the job properly. Consequently, when there is a major accident, the State must often rely on another State, most often the US, France, UK or Canada. While these States (and several others) are willing to help whenever possible, especially on the big major accidents, it is not very comforting from a global safety perspective to have only a small number of authorities with these capabilities, given that a tremendous amount of safety value can be extracted from the analysis of so-called "minor" accidents, and incidents.

If States were to begin taking their recorders from all those accidents and incidents to the handful of authorities with expertise, it would completely overload their capacity. Since this is not happening, we can only assume that authorities are not looking at this data that is readily available, and this is certainly not in the best interest of safety. I would urge those States that do not have expertise to become involved and make it a priority to develop and improve capability in this area. The flight recorders are a key component of a modern accident investigation. If you do not have the capability to work with the data, or miss important safety issues, you are not in a very good position to develop the procedures, training programs or technical changes that will address the safety deficiencies revealed by the flight recorder data. You may not see the trends in operating performance that often lead to a serious incident or accident, especially on new, high-technology aircraft in line operations.

On the subject of FDRs and data, there is the ongoing issue of parameters recorded. As most of you know, there are precise regulations regarding the minimum number of parameters required to be recorded by various regulatory bodies around the world. These rules cater to a diverse range of aircraft and often reflect the lowest common denominator for the applicable category. Investigators always want more data recorded on the FDR but are tempered by the regulation process (which is slow) and must respect legitimate cost issues. As the trend worldwide continues whereby more and more operators are monitoring flight data on a routine basis to identify unsafe trends, a process you all know as FOQA or flight operations quality assurance, airlines are beginning to record on Quick Access Recorders more data than the mandatory requirements. QARs were developed because early-generation FDRs were not readily accessible and it was time-consuming and expensive to extract the data from them on a regular basis. What we have today is two parallel systems, one that survives the accident with the mandatory parameters recorded and one that often doesn't survive, (the later often with greater magnitudes of data recorded.) In fact, the SR111 QAR recorded six times the volume of data that was available on the FDR but unfortunately the QAR did not survive the accident. The evolution of two parallel systems doing essentially the same thing is historical and it is time to develop an integrated system that satisfies both the operators' daily desire for data to operate their fleet safely and the accident investigators' desire to understand what happened when the aviation system fails.

The US-led Flight Data Collection Committee was convened to develop a plan for flight recorders in the year 2015. Is it philosophically acceptable to record more data for a FOQA program that looks at many events in minor detail than for an accident investigation, which looks at one event in excruciating detail? Industry and governments should work together with the objective that all additional flight data beyond the minimum mandatory list that an airline wishes to capture for its FOQA program be available to the accident investigation community in the reconstruction of an accident.

Finally, I would like to touch on the sensitive subject of video or image recording in the cockpit for the purposes of accident investigation. The importance of protection of sensitive voice and potential image recordings is a significant and continually challenging issue; however, we must not lose sight of our overall objective to have the safest possible worldwide aviation system. While we all deplore any misuse of cockpit voice recorders, the definitions of misuse are not universally accepted, and the occasional misuse does not negate all of the significant benefits that CVRs have brought to the worldwide safety community.

In order to successfully introduce image recordings, additional initiatives need to be undertaken to find ways to mitigate or eliminate the airing of sensitive recordings by the media. One potential avenue worth considering is to have CVR and image recordings that have been used for an official investigation purpose "owned" by an international or independent agency (perhaps to an ICAO standard.) The international community could effectively "copyright" the recordings which would prohibit broadcasting with significant financial penalties for copyright infringement. If the National Football League can prevent one of its games from being broadcast without its expressed written permission, then maybe we ought to be able to achieve the same for CVRs.

The personal privacy of the crew must be protected. The public should not have the right to information that should be considered privileged.

Now it's time to come back down to earth, for our approach and landing. I have given you a few issues of an international flavour that lend themselves to the nature of ISASI. As a world-leading venue for advancing aviation safety through occurrence investigation, I encourage you to give these broad issues serious thought, to consider ways to advance them, and ultimately to solve them.

The global community must have the highest level of confidence that we manage the safest transportation system possible.     - That public confidence was shaken last week.

Thank you, and enjoy your time in Victoria.