Canadian Accident Investigations Take Too Long, Survey Says

Vital safety messages, recommendations diluted with passage of time
The Transportation Safety Board (TSB) of Canada has been sharply criticized for taking too long to complete major accident investigations, according to a survey of officials in that country's transportation industry.

"With only a few exceptions, everyone complained about the length of time between an investigation and the availability of the occurrence report," said a summary of the responses from industry "stakeholders" in Canada.

The TSB was perceived as carrying out its mandate, but taking so long that some respondents deemed the timeliness of its work products "to be totally unacceptable."

Canadian-based Sage Research Corp. conducted the survey and covered all modes of transportation in which the TSB is involved: air, marine, rail and pipeline.

To be sure, some accidents are more complicated than others, not only to unravel, but also because of the numerous and often interrelated safety issues involved. The fatal 1998 crash of Swissair Flight 111 stands as a case in point. The $56 million investigation took more than four and a half years to complete, consuming on average more than $1 million per month (see ASW, April 7). A sampling of 10 TSB accident investigations covering a range of causes and a variety of aircraft types shows an average time-to- completion of about 22 months.

The Sage report may be a window into a worldwide problem. Virtually all accident investigation bodies are marginally resourced, with lean staffing levels the prevailing norm. A 1999 Rand Corp. study of the U.S. National Transportation Safety Board (NTSB) reads in many respects like a parallel product of the Sage Survey of the situation in Canada (see ASW, Dec. 13, 1999). Respondents to the Sage survey perceived many strengths and weaknesses in the TSB's operation, but lack of timeliness led the list of perceived weaknesses.

As of October 22, the TSB has 71 ongoing aviation accident investigations, of which 28 are more than a year old. This workload is being carried by 56 TSB aviation accident investigators. That number reflects two recent hires to beef up the staff and to help reduce the backlog of active investigations.

The TSB was invited to respond to a number of key issues raised in the 98-page Sage report. Those responses follow:

1)ASW: Did the TSB commission this independent assessment and, if so, why?

TSB: Yes. The objective of the project was to develop a comprehensive understanding of our stakeholders' needs related to our products and services. The study confirmed that stakeholders do appreciate the contribution TSB makes and the value of receiving our reports.

2) ASW: One of the universal complaints from the stakeholder survey was the perceived excessively long time between the accident and final report. Swissair 111 took more than four years to complete. In the U.S., National Transportation Safety Board Chairman Ellen Engleman has set as a goal that major accident investigations should be completed in two years (see ASW, Sept. 15). How might the time to complete some of these investigations be accelerated? Is there an intrinsic tension between pushing to completion at a sacrifice in depth? Recall the aphorism, "If you want it bad, you get it bad."

TSB: We are committed to producing complete and professional reports on transportation occurrences in a timely manner. Having said that, we will take the time needed to ensure a thorough and meaningful final report. However, our primary interest is getting the safety message out as quickly as possible to those who can best act on the information. By way of example, during the course of the Swissair 111 investigation, 14 recommendations were issued by the TSB, six of these in the first year of the investigation and prior to the publication of the final report.

With respect to improving the timeliness of our investigation reports, the TSB has been reviewing its processes to identify opportunities for streamlining. To achieve that, we have introduced training in areas such as investigation management and communicating safety deficiencies. We are exploring other options for getting our safety messages out.

3) ASW: Stakeholders complained that the TSB has stopped working on safety studies and providing safety workshops.

TSB: We are undertaking a complete review of the role of safety issues investigations.

The TSB continues to offer training to other agencies in such subjects as safety investigations and the investigation of human factors, as time and circumstances permit.

4) ASW: The TSB was criticized for not marketing its products and services more effectively. How do you respond?

TSB: This is important and valuable feedback. The TSB is committed to increasing the level of awareness of our products and services. As well, the TSB is working to enhance the availability of safety information through the Internet.

5) ASW: Some stakeholders said the TSB focuses on large, regulated carriers and accidents involving loss of life and substantial damage, to the detriment of investigating cases where no lives were lost or significant damage occurred, but which have significant safety implications. How do you prioritize your resources?

TSB: Approximately 4,000 transportation occurrences are reported to the TSB each year in accordance with reporting requirements. Practical considerations dictate that only a small proportion of these be investigated. But most reported occurrences by themselves offer little scope for adding to the board's knowledge of the underlying safety deficiencies. Decisions to investigate occurrences are based on this policy, which is available on the TSB's web site at: http://www.tsb.gc.ca/en/common/policies/occurrences.asp.

6) ASW: Some respondents thought the TSB should more actively investigate issues, even in the absence of occurrence(s), to identify latent threats to safety. Do you have the resources for this, and if you did, should you?

TSB: To date, our safety issues investigations have been based on trends or issues arising from occurrences reported to us. All safety issues that come to our attention are assessed and a decision to investigate a particular issue is based on the criteria of our occurrence classification policy.

7) ASW: Occurrence reports were criticized for not being clear about the steps/actions needed to plug identified gaps/lapses in safety. How do you respond?

TSB: The TSB, as part of a thorough investigation process, makes public its findings as to causes and contributing factors. In addition, when warranted, safety recommendations are made directly to those change agents who are in the best position to make the required changes.

8) ASW: Numerous stakeholders lamented the lack of a usable, searchable accident/incident database.

TSB: A searchable web-based database has been developed by the TSB, and we are currently in the process of readying it for public use.

9) ASW: Complaints were raised that the TSB is not investigating accidents that occur outside Canada. The Air Transat A330 dead stick landing in the Azores comes to mind, which the Portuguese are investigating. What are the ground rules governing TSB investigations of Canadian-registered aircraft involved in accidents/incidents overseas?

TSB: The TSB is the government department responsible for ensuring that Canadian safety interests are represented in investigations conducted by foreign governments involving Canadian operators, aircraft, regulators and products. The responsibility is mandated by the provisions in the International Civil Aviation Organization's Annex 13.

The TSB has some involvement with approximately 100 foreign investigations per year, although our investigators only deploy to approximately eight international investigations per year.

Regarding the investigation into the Air Transat fuel-starvation event of August 23, 2001, the TSB contributed significantly to the investigation and now awaits the release of the final report by the government of Portugal.

10) ASW: Of recommendations that are generated, some stakeholders said that not all are linked to findings of safety deficiencies. Should they be?

TSB: Policy requires that all TSB recommendations be based directly on validated safety deficiencies.

11) ASW: With the "revolving door" of people moving from Transport Canada to TSB, some respondents questioned the independence of the TSB. Address briefly the independence issue.

TSB: The TSB is an independent agency that reports directly to Parliament through the president of the Privy Council. The TSB does not report to the Minister of Transport. While staff do move between government departments, this enriches the capabilities of government organizations and in no way compromises the independence of the TSB.

12) ASW: It was suggested that the TSB schedule public hearings to bring out key issues and to advance safety. Does the TSB's enabling charter allow for such public hearings and what is the TSB's position on convening them?

TSB: Section 21 of the TSB Act provides for public inquiries. They would be conducted in accordance with our occurrence classification policy, which identifies a number of considerations, including: (a) whether an inquiry would uncover facts that might not otherwise be made known; or (b) whether an inquiry would result in quicker remedial action.

To date, the TSB has not deemed it necessary to use the vehicle of a public inquiry to meet its mandate.

Stakeholders' Perceptions of the TSB
Perceived strengths:

Professionalism of staff.
Transparency of the investigative process.
Independence from the regulatory body, Transport Canada.
Perceived weaknesses:

Inordinate length of time to complete accident investigations.
TSB takes its independence too far, isolating itself from industry.
Shortcomings, gaps in accident/incident database categorized by aircraft, engine type & causal factors.
Unclear investigative criteria (incidents not involving loss of life or significant damage not investigated, yet these may have significant safety implications).
Inadequate resources (TSB only tends to fully investigate "spectacular" accidents like Swissair).
Recommendations not always linked to findings.
Some evidence (from recommendations) that the TSB operates on an unrealistic assessment of risk tolerance (e.g., zero) vs. regulators and the industry's acceptance of a reasonable level of risk.
Poor marketing of work products and services.
Need for more work on macro safety studies that transcend the particulars of individual accidents.
Source: Sage