"Data and Trust"
Peggy Gilligan, Salt Lake City, UT
May 17, 2016
Western Electricity Coordinating Council
Remarks as Delivered
Thank you, Taud [Olson]. It’s good to be here. I notice that your agenda focuses on the characteristics of organizations known for high reliability. Keeping your eye on the ball is part of that, and in our case, the ball is the stakeholder—from the passenger to the aviation industry.
We go out of our way to make situational awareness a 24/7 routine. As part of that, my management team and I get a text message anytime something happens in the system—not just accidents or crashes. If there’s an unruly passenger, I get a text. If an aircraft diverts because of bad weather, I get a text. If there’s a ground stop in Newark, I get a text. As you can imagine, I get a lot of texts. Some of those apply directly to my area of responsibility, some don’t. But each of them is part of the larger picture—the bigger perspective—that we all must have to keep things moving.
In addition to those texts, every Monday, Wednesday and Friday, I meet with my senior management team. At the top of that meeting, I get a briefing from our Accident Investigation and Prevention group about what’s happened in the system—every accident, every incident over the previous couple of days. This gets to my point about remembering—really remembering—where our focus needs to be. When there’s an accident, I’m told about the aircraft, make, model, and registration, the weather conditions, whether or not it was being guided by air traffic control … the basics. And, the briefing always includes information about the pilot: experience, flight hours, ratings, age, etc. It’s not: an accident happened, next subject. It’s recognition that people were involved in an accident.
They’re the husbands, wives, fathers, mothers, sons, daughters, sisters, brothers, friends of the people who place their trust in the FAA to put a safe system in place.
I can never forget that, and I’m quite sure you believe the same. Our missions are very similar, and people are at the core of both of them.
You’ve asked me to talk about reliability from the perspective of the Federal Aviation Administration, and I’m happy to do that. In our society, there are some things we just expect to be reliable: those who provide electricity. I’m guessing you knew that one. Your surgeon. Your dentist. The guy who does your brakes. The guy who does your taxes.
And commercial aviation is on that list, too. Keep this date in mind: February 12, 2009. I’ll get back to the importance of that in just a moment. In U.S. commercial aviation, there are 50,000 flights every day and But back to February 12, 2009. That’s the date of the last passenger fatality on a commercial airliner in the United States of America. It’s hard to imagine, but it’s true. Since that day, 7-1/2 years ago, 5 billion people have flown in our system—without a single passenger fatality.
Why? Because the system is so reliable. How did we get to that state? The answer to that question goes back to our history—right up to the mid-1990s.
The post-war 1950s gave birth to the airline industry … and about a crash a week. In 1956, a TWA Lockheed L-1049 and a United Airlines DC-7 collided not far from here over the Grand Canyon. 128 dead. No survivors. That was the first aviation accident to claim more than a hundred lives. As a result, industry itself requested to be regulated and demanded better air traffic control services. Congress responded by creating the Federal Aviation Agency.
The disasters that followed showed that more needed to be done. A DC-10 lost an engine in Chicago in 1979. Three years later, an Air Florida jet crashed into the icy Potomac less than a mile from FAA headquarters. In 1988, an Aloha Airlines jet literally came apart in midair. Headlines began to call us the “tombstone agency”—that we only made safety improvements after catastrophic accidents where many lives were lost. And how many of you remember that you used to be able to buy an insurance policy right at the gate? Perhaps that was not the best way to instill confidence in the flying public.
We were making safety improvements through all those years, but then came ValuJet in the Everglades, and TWA 800 crashed in New York. Suddenly—but maybe not suddenly enough—the need for change became a mandate for change, and the people leading the charge were pointing their finger right at the FAA.
The ideas on how and what to change were numerous. We had literally hundreds of recommendations on the table suggesting ways to fix things. I think the push for reliability came when a Presidential Commission called for a 5-fold reduction in fatal accidents … in 10 years.
Take a look at this next regarding onboard fatalities. There’s a lot of information there, but I want you to focus on the green peaks and the time line across the bottom. That’s at the heart of this story.
Utilizing a government/industry group known as the Commercial Aviation Safety Team, we did it. We pulled together the collective brainpower of safety experts from FAA, NASA and DoD. Industry experts included manufacturers, pilots and airlines. Names you know, like Boeing and Airbus… United, Delta and American. And employee unions representing pilots, flight attendants and controllers.
The first big step was the development of a prioritized safety agenda. Because, as we all know, when you try to keep your eye on everything, very quickly, you end up keeping your eye on nothing. The Commercial Aviation Safety Team developed and implemented a safety portfolio targeted at the high fatality risk areas. At a time where budgets are shrinking and staffing is critical, risk-based decision-making just makes smart business sense. We put our time, money and people where we believed we would get the biggest return on the investment.
Let’s go to the next slide … U.S. Air Carrier Fatalities. Those peaks from the previous slide are the green line you see, except this is a look after the work of the Commercial Aviation Safety Team. We’ll talk in a moment about what it takes to keep that line essentially flat, but suffice it to say that the trend very clearly is heading in the right direction.
Risk areas are a moving target. Given the speed with which industry is developing new technology—drones and commercial space rockets come to mind—we continually have to evolve with it. The speed that unmanned aircraft entered our airspace required an entirely new approach to safety rules. First off, we’d never dealt with an aircraft that weighed the equivalent of two sticks of butter. And in our experience, pilots were, well, pilots. With drones, you become a pilot because of something Santa Claus left under the tree. We’re not used to that. Count yourselves lucky that Brookstone does not sell power grids. We had to register better than 400,000 unmanned aircraft owners in a matter of months. Typically, rulemaking takes years. Saying that we’ve had to evolve puts it mildly.
We’ve had to keep our data collection efforts state of the art as well. In order for this to work, there has to be a constant flow of knowledge back-and-forth from industry to government and vice versa. Knowledge transfer and human performance—at micro and macro levels—is key to this. It has to happen person-to-person and it has to happen system-to-system.
As we became more focused on the need to mitigate risk and increase safety, we developed means and mechanisms to collect and share data from everyone and everything that touches the airplane. Rows and rows of ones and zeroes, piles of voluntary inputs from firsthand experience.
We filter those through our Aviation Safety Information Analysis and Sharing System. Those results get racked and stacked and sent to the Commercial Aviation Safety Team. What you have then is a roomful of people who have a decided interest in making everyone as safe as they can be. There’s no competitive advantage in safety because they’re all sharing the same sky. Commercial aviation recognizes this and is all too willing to share safety information.
The key to its success is the cooperation of government and industry, and the voluntary adoption of proposed safety enhancements. This would not have been possible through regulatory mandates.
Through this process, we’ve eliminated accident causes and risks like Controlled Flight into Terrain. We now have terrain awareness warning systems that have helped us virtually eliminate CFIT. We’ve also put in place safety enhancements targeting loss of control of the aircraft. We’ve put navigation procedures in place that were targeted to reduce the risk of midair collisions, especially on arrival and departure. If you think about it, airports have the most congested airspace and the highest risk for a mid-air collision.
Not every safety improvement has come through CAST. Let me give you an example of a safety advancement that came about through a different path.
Please head to the next slide on Inflight Engine Shutdown Rates … it’s an eye chart … but before you start looking at that, you’ll need context. Let me start by asking for a show of hands: how many of you have flown across an ocean? This is not for a show of hands. When you were looking out the window, did you ever think about the reliability of those engines?
Well, we do, and we know that an aircraft engine is one of the most reliable systems on the planet. Commercial aircraft engines are meticulously maintained by a group of maintenance professionals who take the word meticulous to a new level.
For a long time, most twin engine jets had to fly within 60 minutes of a suitable airport. For overwater flights, that bends routes that could have been a straight line. Over time and based on decades of data, we changed the rules to allow commercial airliners with two-engines to be certified for extended operations up to the limits of their capabilities and engine reliability. That effectively removed all limitations on twin engine flying.
The “X” on the slide are the engines designed in the 1960s that you found on DC-8s and the early 747 and A-300. The green squares are the second generation engines—designed in the 80s—that’s the 767, the MD-11 and the A-340. The purple triangles are planes like the 777 and the A-380. The takeaway from this chart is obvious: Most pilots probably will never experience the shutdown of an engine in their entire flying career. That Y axis on the left shows shutdowns per thousand hours as it drops from 1 to point one to point oh-one … to point zero zero one. It’s a staggering representation of reliability.
So … what does this mean to you? For starters, your aircraft can be as much as 2,500 miles away from land and still operate on one engine. This saves flight time, which saves fuel, which saves money, and is good for the environment. Thirty years ago, this would have affected only about a thousand flights in any given month. Now, it’s upward of a thousand flights per day. Engine shutdown rates have gotten lower and lower over the years as manufacturing and maintenance practices have gotten better and better. And that data allowed us to make system improvements.
But data alone is not the answer. How you obtain the data, and what you do with the data are as important as the data itself. We push for voluntary submission of data. Think if you will about running a stop sign and then driving to the police station to report yourself. That’s just not how it’s done, right? But think about it another way. Let’s say five different drivers show up at that police station in a week and report the same thing. And then the police take a look at the intersection, and they discover a tree has occluded the sign. Now, let’s take it to the next level, let’s say this is Northern Virginia, and the tree that occludes the sign is the state tree: the dogwood, and dogwoods are planted by the commonwealth as part of roadway beautification efforts when new signage is installed. And then, you find out similar scenarios are unfolding in southern Virginia.
All of a sudden, the lone “violation” turns out to be a systemic safety problem. It’s not the car. It’s not the driver. It’s not the sign. It’s the tree. But if you just write a ticket to the first guy, you’ll never know that.
Let me give you an actual example. One airline had uncovered an issue with the installation of a washer on a landing gear assembly. The washer we’re talking about is located between the outer wheel bearing and the axle nut. You need the washer so that the wheel can be tightened.
To make a long story short, sometimes, the washer stuck to the grease on the wheel during disassembly … and sure enough, it was easy to overlook on assembly … especially because not every assembly required this particular washer.
The glitch turned up in our voluntary safety reporting system. Turns out the issue wasn’t as infrequent as everyone thought. The same situation was occurring elsewhere. Because of voluntary reporting—with no fear of retribution—we found it, and corrected it. This information was shared with all of the other airlines to manage the risk. It became an accident that never had the chance to happen.
When all of the data comes in voluntarily, and the people in the system know that they don’t have to fear hardline enforcement for unintentional mistakes, the data stream thickens and trends emerge. And hiding in those trends are the risks … the precursors of accidents that we can eliminate before the accident occurs.
It’s natural for you to ask, “What about bad actors?” This approach to safety and compliance is not a safe haven for the unsafe. If you repeatedly break the rules, or if you break a rule that shows complete disregard for safety, you will be brought to task. Our desire to get voluntary data is by no means a get out of jail free card. But in a system as complex as the worldwide aviation system—where we know we have risks we haven’t address—we need insight into just what’s going on.
We have a lot of folks who hold our certificates—about 80 airlines, 736 pilot training schools, 4,700 repair stations, 18,000 commercial aircraft, 290,000 general aviation aircraft, 400,000 drone owners, 720,000 active pilots and 100,000 flight instructors.
When you’re dealing with a system that big, with that many moving parts, it takes something more than data and a willingness to share. It takes trust, the kind of trust that’s borne of the regulator and the regulated working toward a common goal. I’ve mentioned the trust on the part of the pilot … consider the trust on the part of the FAA safety inspector. His job is to make sure you follow the rules. Now, that’s changing. The inspector has to trust the system and the people who use it. The focus is on compliance and education.
That takes steps on the part of the senior leadership to keep reemphasizing the point that we’re all in this together. So when we go out into the field and meets front line employees, we must reinforce that this is about compliance, not enforcement. Our inspectors are very good at what they do. These people are veteran pilots, maintenance workers … from every and any walk of life in the aviation industry. They bring as much experience into the cockpit as the Captain or the first officer. We’re talking about a level of knowledge that’s a mighty deep pool.
Developing trust with professionals like this takes more than a poster and a slogan.
Our approach to this is employee engagement. We hold national town hall meetings with employees once a quarter. They’re broadcast live via satellite. We hold virtual brown bag lunches with remote offices once a month. My entire executive team of 18 goes out in teams of two to visit our regional and field locations—40 different sessions in a year.
That’s how we build trust in our workforce of 7,400.
Outside, we facilitate a bi-annual closed-door meeting called Aviation Safety InfoShare—invitation only. No press. All of the airlines and industry personnel get together to wade through safety issues and safety trends. Almost 800 people attended last time in Philadelphia. They know at these meetings, no subject is taboo, nothing is off-limits for safety’s sake. But they also know that none of the information is used against them—it’s all for safety.
In closing, let me say what pleases me most about this opportunity to speak is that your business does not have a direct link to aviation—but you’re looking outside yourselves to see what you can learn and what you might apply. Just last month, we spoke to automobile manufacturers who had the very same intent: to learn what they could from what we’ve been able to do. So I hope this was helpful. If you have questions, I’m happy to answer them. Thank you.