Thanks to the National Automotive Dealers Association for inviting me to your annual convention to speak on behalf of the National Transportation Safety Board. The NTSB has developed a good working relationship with NADA, and we particularly appreciated your very helpful participation in our recent conference, “Reaching Zero Crashes: A Dialogue on the Role of Advanced Driver Assistance Programs.” We are also pleased to see that many of your dealerships are taking active roles in “Senior Promise” this prom season, to address the problem of the disproportionate number of teen fatalities from crashes on our streets and highways.
Today I would like to speak about driverless cars, some of the issues associated with bringing automation into motor vehicles, and how the NTSB can help with that process. The automobile industry has exploited IT advances for decades to mitigate injuries from crashes, with improvements in seat belts, air bags, and passenger compartment integrity. Those efforts have undoubtedly saved thousands of lives a year.
In recent years the industry has begun to exploit IT advances to prevent crashes from occurring in the first place, and that provides the opportunity to save tens of thousands of lives a year.
Having said that, the journey will be very challenging, as automation advances in other industries have been. Today I will discuss how we can make that journey less challenging by learning from other automation efforts, such as in aviation. I will also discuss how dealers can help, because dealers will play a major role in moving toward automation.
To put my remarks in context, I would like to describe what the NTSB does. The NTSB is an independent federal agency that investigates accidents in all modes of transportation to determine what caused them and to make recommendations to prevent them from happening again. We are not a regulator. Instead, our primary product is recommendations, and we send them to any entity that can improve safety, including manufacturers, state governments, and NHTSA. Our world-class investigators and analysts don’t like to give up until they have the answer, and the recommendations that they create are so compelling that the recipients respond favorably more than 80% of the time, even though they are not required to. We like to think that the implementation of our recommendations has helped to make transportation safer for all of us.
I describe the NTSB as “independent” because of the structure that Congress gave us, to its credit. The agency is led by five Members who are nominated by the President and confirmed by the Senate. The most important aspect of our independence is that the Members serve fixed terms, and the terms are staggered. Most political appointees serve at the pleasure of the President. In the real world, that means that if the appointee does something that is politically challenging or unpopular, the appointee may be out of a job. Serving in fixed terms helps to insulate the Members from being removed by a President who has been lobbied by a manufacturer, an operator, or a union that is dissatisfied with our investigation of an accident that they are involved in.
In addition, very few political appointees have a substantive knowledge requirement. Our enabling statute, on the other hand, requires that at least three of the five of us have some relevant expertise. Moreover, the statute helps to create party balance by permitting only three of the five of us to be of the same political party as the President.
The purpose of these requirements is to help ensure that our determinations of cause and our recommendations come from the facts and the evidence of our investigations, rather than from political forces or lobbying. In my view, the structure that Congress gave us does this very well.
My remarks today come from the context of our decades of experience as accident investigators. We only investigate a small fraction of highway crashes, typically more systemic ones involving commercial or transit vehicles, such as tanker trucks, motor coaches, and school buses. We rarely investigate passenger vehicle crashes, but we are investigating the first fatal crash of a Tesla that was being driven in an automated mode in Florida last May, because of its potential importance to the introduction of automation into motor vehicles.
We have a Most Wanted List of safety improvements that have been particularly challenging and need increased promotion and emphasis. One MWL item I’ll discuss today, because it relates to automation and because dealers can play a role, is “Increase Implementation of Collision Avoidance Technologies”
Driverless cars are coming, and their potential for improvement is amazing. First and foremost, driverless cars could save many, if not most, of the tens of thousands of lives that are lost every year on our streets and highways – a very tragic and unacceptable number that has been decreasing for several years but has recently taken a turn in the wrong direction.
Driverless cars could also increase the amount of traffic that our roads can safely carry because, instead of maintaining a car length separation for every 10 mph, as I’m sure we all do, driverless cars could reduce that separation. Stay tuned for what other changes might be possible.
How might that happen? Ideally, with automation.
It has been said that more than 90% of the crashes on our roads are due to driver error. The theory of driverless cars is that if there is no driver, there will be no driver error. Ideally, removing the driver would address at least four issues on the NTSB’s Most Wanted List – fatigue, distractions, impairment, and fitness for duty. The automation in driverless cars would presumably also address a fifth item on our list that I’ve already mentioned, namely, improved collision avoidance technologies.
Extensive experience in a variety of contexts has demonstrated that automation can improve safety, reliability, productivity, and efficiency. That experience has also demonstrated that there can be a downside. As noted by Prof. James Reason, who is a world-renowned expert in complex human-centric systems:
In their efforts to compensate for the unreliability of human performance, the designers of automated control systems have unwittingly created opportunities for new error types that can be even more serious than those they were seeking to avoid.
Our investigation experience provides three lessons learned that support Prof. Reason’s statement. The first is that the theory of removing human error by removing the human assumes that the automation is working as designed; so the question is what if the automation fails. Will it fail in a way that is safe? If it cannot be guaranteed to fail safe, will the operator be aware of the failure in a timely manner, and will the operator then be able to take over to avoid a crash?
An example of the automation failing without the operator’s timely knowledge occurred in Washington in 2009 in a subway crash that tragically killed the train operator and 8 passengers. In that accident, a train temporarily became electronically invisible, whereupon the symbol of the train disappeared from the display board in the dispatch center. When a train became invisible on the board, an alarm sounded. This alarm, however, sounded several hundred times a day, so it was largely ignored.
Unfortunately, when the train became electronically invisible, there was no alarm in the train behind it regarding the electronic disappearance of the preceding train. Thus, the operator of the train behind was unaware of it. Instead, based upon the electronically unoccupied track ahead, the automation in the train behind began accelerating to the maximum speed for the area. By the time the operator saw the stopped train and applied the emergency brake after coming around a curve – which limited her sight distance – it was too late.
Another lesson learned in support of Prof. Reason’s statement is that even if the operator is removed from the loop, humans are still involved in designing, manufacturing, and maintaining the vehicles, as well as the streets and highways they use. Each of these points of human engagement presents opportunities for human error. Moreover, human error in these steps is likely to be more systemic in its effect – possibly involving several vehicles – and more difficult to find and correct. An example of this lesson learned is the collision of an automated – driverless – people mover into a stopped people mover at Miami International Airport in 2008. That collision was caused largely by improper maintenance.
The most fundamental lesson learned from our accident investigation experience in support of Prof. Reason’s statement is that introducing automation into complex human-centric systems can be very challenging. Most of the systems we have investigated are becoming increasingly automated but are not fully automated. As a result, we have seen that the challenges can be even more difficult in a system that still has substantial human operator involvement and is not completely automated.
Given these lessons learned, the question is how can we bring automation safety onto our streets and highways. That process involves several steps.
The first step is that the automation must be well designed. Good design is the foundation of safety improvement, but it is in the province of the manufacturers rather than the dealers.
The manufacturers will not need to re-invent the wheel because there is much to learn about automation from others who have done it in many industries for decades. In transportation, the leader is aviation, but despite decades of experience, and despite intensive and recurrent training of pilots, they still have automation accidents. We can inform the automation process with our experience investigating accidents resulting from things that went wrong with automation.I’ve already seen a need for the auto manufacturers to learn from aviation’s experience. At the Detroit auto show recently, then NHTSA Administrator Rosekind noted that the industry needs to focus more on human factors. In aviation, they discovered that automating something did not necessarily make it safer, so they evolved from “automate it because we can,” i.e., because it is technologically feasible, to “automate it only if it makes the person-machine system work better.”
The next step is for all of you dealers to encourage your customers to purchase the automation. Collision avoidance systems, for example, are already widely available, and hopefully dealers will encourage customers to purchase them. We showed the safety benefit of these systems by issuing a Special Investigation Report, or SIR, in 2015, on forward collision avoidance systems, or CAS, which consists of a collision warning system accompanied with autonomous emergency braking (AEB). We recommended that NHTSA mandate these systems as standard equipment, not just high-end options. Rather than mandating the systems, NHTSA reached voluntary agreement with most automakers to make AEB standard by 2022. Until then, we encouraged people to buy them by issuing a Safety Alert to drivers and fleet owners, making them aware of the safety value of CAS.
We also hope to encourage consumers to buy them by recommending, in our Collision Avoidance SIR, that NHTSA develop testing protocols and rating criteria for collision avoidance systems; incorporate those ratings into the 5-star safety rating scale; and disclose the performance of the collision avoidance system on the vehicle’s Monroney label. In response to our recommendation, NHTSA plans to revise its New Car Assessment Program (NCAP) by 2018, including CAS and many other driver assistance systems, and these changes will be reflected on Monroney labels beginning with the 2019 model year.
Researchers at the University of Iowa found that the more drivers knew about five driver assistance technologies – adaptive cruise control, blind spot monitor, lane keeping assist, parallel parking assist, and rear cross traffic alert – the more they found them useful, the more they trusted them, and the less apprehensive they were about them. Dealers are uniquely positioned not only to alleviate customer concerns, but also to show customers how these systems improve their safety and reduce their costs by reducing the likelihood of crashes.
In commercial vehicles, we recommended that NHTSA develop ways to assess forward collision warning and avoidance systems in commercial vehicles, develop standards for stability control, and when these standards are completed, require stability control for vehicles over 10,000 pounds.
The next step in the process, in which the dealers play a major role, is showing the customers who decide to purchase the automated systems how to use them. The design of the automation is never perfect, so dealers will play a crucial role to ensure customers know how to use it. Dealers demonstrated their willingness and ability to educate their customers with car seats by hosting child safety seat fittings. If customers buy collision avoidance, it is important to make sure they are comfortable with what it is and how to use it.
In order to do this, the dealer sales staff must be trained. Dealers will encounter a wide range of customers – from young to old, from tech-savvy to tech-fearing – and teaching customers will be a major challenge, especially If the only face-to-face training the customers get is when they purchase the vehicle. The objective is to get customers comfortable with what the automation does and does not do, thereby reducing the likelihood that they will turn it off because they don’t understand it or because they were startled or confused by it.
Last year at this conference, NADA announced a partnership with the National Safety Council and the University of Iowa to promote MyCarDoesWhat to give customers a source of information on their laptops – once again showing NADA’s willingness to embrace safety. However, several studies, including a recent MIT study, suggest that understanding of safety features is less than perfect at the level of the salesperson, and that consumers are still confused about what various driver assistance technologies do. The challenges for dealers are improving the training of their staffs, and bridging the knowledge gap for their customers.
Last but not least, the dealers can play an important role by providing feedback to the manufacturers regarding what the automation is doing well and what it is not. Many customers will report their problems to their dealer rather than directly to the manufacturer, and if something goes wrong with the systems and the car is brought to the dealer for repair, your maintenance staff will be a valuable source of information to the manufacturers. We have found throughout transportation that front-line reporting of what works well and what needs improvement is critical for effectively and efficiently improving systems that automate operational tasks.
In conclusion, more automation of the driving task is coming – one day we will even see autonomous vehicles, but they will not dominate our passenger vehicle fleet for a very long time. For now, people will purchase vehicles in which automation assists human drivers. We know from experience that automation can greatly improve safety, but that the human-automation interaction is a prime area for new accident causes. As automation increases in passenger vehicles, dealerships are extremely important. Nobody else can play the same role in encouraging the purchase of driver assist features, showing customers how they work, and pointing customers to reliable educational resources.
The challenge is not only keeping up with new features, but also recognizing that these features demand a new level of attention to customers to ensure that they understand the features, and compiling and relaying feedback to the manufacturers about new driver assist features to help the manufacturers improve their products.
Finally, I cannot overstate the importance of forward collision avoidance technologies in making highway transportation safer and in providing the foundation for driverless cars. Working together, we can make the transition to more automated vehicles – with their promise of enormous safety benefits – more smoothly and safely.
Thanks again for inviting me to speak today. I would be happy to take any questions.