Keynote address to the Northern Air Transport Association
Annual General Meeting

Kathy Fox
Chair, Transportation Safety Board of Canada
Whitehorse, YT, 26 April 2016

Check against delivery.

Thank you very much for the opportunity to speak to you today. It's a pleasure to be here once again—and on your 40th anniversary. Congratulations!

Jon Lee has already given his excellent presentation on accidents and occurrences “north of 60” and included an update on where things stand with respect to the Safety Issues Investigation on the risks associated with air taxis, so I won't be covering what he has done already. Instead I'd like to share some of what took place during an event held last week in Ottawa—where the TSB hosted the first-ever multi-modal transportation safety summit that brought together a group of senior executives from across Canada's transportation industry, along with representatives from worker associations and government officials, to exchange ideas about two main issues:

One, the use and sharing of safety data to find out where the real safety risks lie in their organization.

And two, improving the flow of that information.

Neither of these two concepts is new, but their use varies widely depending on the mode of transport, the organization, and even who's in charge. And so the goal of the summit was to share best practices and to see how other industries and organizations have implemented some of these ideas—in short, to see what's working, and to discuss the opportunities and challenges these ideas present.

But why did we choose those two concepts in the first place: the use of safety data, and improving the way it flows?

The short answer? Because most accidents, whether big or small, can be attributed to a breakdown in the way an organization identifies and mitigates hazards and manages risk. For instance, multiple investigations by the TSB have shown examples where a safety issue was not identified in advance—either because the data was not available or it was not being mined. Conversely, we have seen occurrences where issues were identified in advance, but weak (or missing) safety management processes contributed to an inability to take action.

In theory, then, if a company improves the way it identifies hazards—and if it gets the right information to the right people at the right time to mitigate risks—things should be “safer.”

Safety, though, is just one variable for companies, one of the many “top priorities” that senior managers have to constantly juggle. Granted, it's one of the most important—because if you're not “safe,” or not perceived to be safe, you may lose customer and public confidence, have costly accidents, face litigation and eventually be out of business. Still, we acknowledge: there's a balance that's required between safety and production.

The process of an organization finding that balance—of finding its way between operating with no risk (which would only be achievable by not operating at all) and operating with significant risk—has been described as “navigating the safety space”.Footnote 1

And as you are no doubt aware, the people who are ultimately responsible for navigating that space and achieving that balance—by establishing the goals and the priorities, by assigning the resources, and by determining the rules for the operating environment—are those typically higher up in an organization. Management. And their decisions can have a huge impact.

Some companies are clearly better at navigating this safety space than others. Why, though? Well, it's not just the existence of processes. A Safety Management System, on its own, for instance, is not enough. Instead, processes must be supported—and sustained—by leadership. In other words, it's about the safety culture of an organization, and how the safety risk management processes in place are integrated with the culture.

And who determines the culture of an organization? In large part—again—it's the same people who determine its direction and set its priorities: management. Their beliefs, their priorities, and above all: how they behave.

Since good data is critical to good decision-making, then where is the best source of accurate, reliable safety information?

Obviously, some of it comes from listening to your employees. They're on the front lines, as it were. They identify issues and make reports, telling managers what they see and what they are concerned about. And make no mistake, employees are good sources of information. But they don't always catch everything, nor—for a variety of reasons—are they always … forthcoming.

That's what we mean when we talk about information flow. And if it's flowing well, it's because employees feel empowered to share what they know, doing so freely, without fear of reprisal. If not, if there's no “just culture”—if the safety culture is less than robust, and if management is not receptive to hearing when employees observe unsafe practices or unsafe events—then employees will be reluctant to come forward. In which case, the right information doesn't get to the right people at the right time.

And so an additional, maybe bigger source—certainly a less … subjective source—is the recorded data you generate from your own operations.

Currently, manufacturers offer several stand-alone lightweight flight-recording systems. These are relatively inexpensive, and they can record combined aircraft parametric data, cockpit audio data, airborne images and/or data-link messages. And while many aircraft fall outside the requirement set out by CARSFootnote 2 for the mandatory use of flight data recorders and cockpit voice recorders— requirements Transport Canada does not currently intend to extend—I'd like to make the case for exactly that.

For one thing, it's of tremendous value to accident investigators.

In 2013, for instance, the TSB released its report into an accident involving a de Havilland DHC-3 Otter, owned and operated by Black Sheep Aviation, which suffered an inflight-breakup and subsequent crash shortly after departing Mayo, Yukon. Our investigation, however, was ultimately unable to determine the cause for the aircraft's departure from controlled flight, mainly because cockpit or data recordings were not available.

The TSB subsequently issued a recommendationFootnote 3 urging Transport Canada, in part, to “work with industry to remove obstacles to and develop recommended practices for … the installation of lightweight flight recording systems by commercial operators not currently required to carry these systems.”

Now fast forward to 2016—just under a month ago—when a Mitsubishi MU-2B-60 collided with terrain approximately 2 kilometers from the airport in les Îles-de-la-Madeleine, Quebec. This was a high-profile accident, receiving extensive, nationwide coverage. And while the ongoing investigation means I am unable to speak at length about the circumstances, there was an onboard recording device that our investigators were able to recover and our Lab specialists are currently working with the manufacturer to download the data. In the coming weeks, TSB investigators will be analyzing any data that can be successfully extracted and hopefully, it will help us better understand what happened and why.

But beyond the work of the TSB, this kind of data may also be of tremendous value to companies, because inexpensive and lightweight flight-recording-system technology presents an opportunity for flight data monitoring, or FDM. This can allow smaller operators—those who may not be mandated to have an expensive flight data recorder like the larger carriers—to monitor, among other things: compliance with standard operating procedures, pilot decision-making, and adherence to operational limitations.

In fact, developing recommended practices for the implementation of flight data monitoringwas the other half of our recommendation that emerged from the Black Sheep accident.

For one thing, FDM is already well-established with the larger carriers, where companies have used it to look at normal operations and identify problems such as: unstabilized approaches and rushed approaches; exceedance of flap limit speeds; excessive bank angles after takeoff; engine over-temperature events; exceedance of recommended speed thresholds; ground-proximity warning systems (GPWS)/terrain awareness and warning system (TAWS) warnings; onset of stall conditions; excessive rates of rotation; glide path excursions; and vertical acceleration.

FDM has also been implemented in other countries, and it is widely recognized as a cost-effective tool for improving safety—especially as an integral component of a company Safety Management System.

As an aside, it's already being used successfully here in Canada, too.  For example, Phoenix Heli-Flight, a Fort McMurray-based operator and one of the companies that presented at the recent TSB summit, has been public about its use of flight data monitoring and the resulting benefits. When the information is used responsibly—that is, within the context of a non-punitive, just culture—it's possible to get a deeper, more in-depth understanding not just of an accident, but of the underlying issues, such as the reasons for non-compliance with SOPs. Moreover, as prices for this kind of technology fall, operators have told us that there is a very strong business case for using it, and that the equipment eventually pays for itself.

As far as the TSB is concerned, that's a pretty good case for the use of safety data, and it's why we saw it as one of the key issues to be discussed at our safety summit.

But there are some obstacles in the way.

Our Act, for instance, protects voice recordings. They may only be used by us, and only in the context of an occurrence investigation. In other words, these recordings—which can shed valuable light on the causes and contributing factors of an occurrence, and which companies may understandably want access to—may not be used in disciplinary proceedings, or in proceedings relating to the capacity or competence of an officer or employee, or in legal or other proceedings.

Air operators are not the only ones struggling with this issue. In the rail mode, the TSB has long pushed for voice and video recorders in the cab of locomotives. Again, access to this kind of information would be of tremendous benefit to our investigations, and there have been several high-profile rail accidents over the past decade in which the lack of an in-cab voice or video recorder has hampered our efforts to find out exactly what happened and why.

We, obviously, see enhanced data as essential for fully understanding the sequence of events and for examining crew actions and interactions. TC and other stakeholders, meanwhile, see value in expanding the use of on-board recordings for legitimate safety purposes within the context of pro-active SMS activities. But again, we stress that it would have to be a non-punitive approach that doesn't seek to single out and punish employees, but rather seeks to learn from what happened to advance safety for everyone.

Currently, however, there is no requirement for railways to install this kind of equipment in locomotive cabs.

Which is exactly why, last spring, a joint TSB/Transport Canada safety study was started, focusing on Locomotive Voice and Video Recorders (LVVR). This study was carried out in conjunction with key rail stakeholders including CN, CP, VIA Rail, GO transit—as well as Teamsters Canada, the trade union representing the interests of railway operating employees. The primary objective of the study's initial phase was to assess the technology and operational aspects of voice and video recordings, demonstrating on a small scale the safety benefits that could be derived from their use. The study also sought to identify best practices, identify and evaluate implementation issues, and collect the required background information for the future development of legislative, regulatory and rule changes.

In other words, if we're going to do this—going forward, for all transportation modes, not just rail—how will we do it? For example, how will we ensure the information is used by companies in a non-punitive way? What rules will need to be put in place? And what changes will be required to the legislation?

Those are good questions, and the answers aren't necessarily clear—yet. But the LVVR study is just about completed, and our team is now drafting the final report which will be shared with stakeholders in all modes of transportation.

It is our hope that this study will help point the way toward answering those questions, and toward the changes that are going to be required in order to allow all of us—the TSB and operators—to make full use of this kind of data for safety purposes.

I need to say here, however, that even if and when these changes do unfold, and even if companies are able to better access these on-board recordings in a pro-active, non-punitive SMS—even if you have all the information in the world—you still need the information to flow. That is, you still need the right people in the right position to be receptive to what they hear—to heed it and be prepared to act.

And so again we come back to management, and to safety culture.

Because how safety information is received and addressed by the receiver, or more generally, by the company, will make a huge difference in what—and how—information will be transmitted in the future!

A punitive approach, for instance, is limiting to the flow of safety information. In fact, that same punitive approach may not prevent the recurrence of normal human failings that are a part of any system. It may only drive the information underground.

Let me give you an example: if a helmsman turns right instead of left when so directed by the ship's master, running the ship aground, and the resulting investigation determines that both were in a fatigued state as a result of the shipboard schedules, should the helmsman be disciplined for turning the wrong way?

As we go forward, it's questions like these that must still be addressed. No one, after all, wakes up in the morning and says “I think I'll have an accident today”—not ships' masters, not helmsmen, not locomotive engineers, not pilots, and not company executives. And so the approach to safety can't just focus on compliance with rules, or even on a single employee or individual: the approach must be company-wide, with buy-in from everyone—all the way up and all the way down.

And as we go forward, what's going to be needed is a frank and open discussion of exactly how we view safety, and which paradigm we choose to deliver it: a punitive, reactive, regime of individual culpability … or a pro-active, non-punitive, systemic approach that is constantly looking at data, and making sure it flows the way it needs to, in order to adapt, to identify risks and hazards, and put in place measures that help mitigate the risks for everyone.

That's a lot to think about, so I'd like to wrap up my time today with a quick update on something that we at the TSB get asked fairly often, and for which I have a much simpler answer. Where are we on the status of our annual reassessment of all recommendations?

The answer: mostly done. Internally, we've completed our review of all outstanding recommendations, in all modes, not just aviation. We've identified where progress has been made—or not—and if it's enough. The final stage—publishing those responses on our website—is already underway and will be completed over the next few weeks.

That process, in turn, will kick-start our updating of the TSB's safety Watchlist, which is the list of those issues posing the greatest risk to Canada's transportation system. We published our most recent edition of the Watchlist back in 2014, and a lot has happened since then, so there will certainly be some changes. I can't give exact dates yet, but you can expect a new edition, likely with updated or even new issues, before the end of this year.

I'd like to close by congratulating you once again on a milestone anniversary. Forty years is a long time. That's four decades of “promoting a safe and effective Northern air transportation system.” It's 40 years of advocating for “northern air transport positions,” and “establishing and maintaining partnerships within the industry, governments and other interested parties.” To come this far takes a clear vision and mandate, commitment, perseverance, and teamwork. So, well done!

Thank you.


Footnote 1

Reason, J. (1997) Managing the Risks of Organizational Accidents. Ashgate Publishing.

Return to footnote 1 referrer

Footnote 2

CARS 605.33

Return to footnote 2 referrer

Footnote 3

TSB recommendation A13-01

Return to footnote 3 referrer