US car safety organisation plans to promote driver assist systems
The US pioneered independent consumer crash testing back in the 1970s to promote advances in car design that would reduce road deaths and improve safety. Fatalities have fallen consistently ever since, but with 30,000 lives lost each year still remain too high. Adrian Lund is the president of the Insurance Institute for Highway Safety (IIHS), the independent organisation responsible for the crash test programme. Car Safety Rules talked to him about the situation
How happy are you with the progress being made on safety in the US?
We’ve made progress and there’s progress yet to be made. Motor vehicle travel is safer than it’s ever been and safer than just five years ago. We’ve improved vehicles by making car roofs stronger over the last five years.
More recently, we’ve been pleased that automakers are responding quickly to our new small overlap frontal crash test. They are paying attention to the safety issues and they’re anxious to make progress when they figure out what the target is.
We’re also seeing them compete to fit crash avoidance technology, which is also a good thing. We don’t fully understand which of those technologies are really going to be effective in reducing crashes yet, but some of them are showing promise.
We’ve also made progress in this country’s laws in trying to affect driver behaviour. Probably the biggest thing is that we have an increasing number of states that have enacted First Offender Alcohol Interlock laws. When we do manage to arrest and convict people who are driving over the alcohol limit, even if they’re a first offender, to get their licence back they have to install an alcohol interlock (an in-car breathalyser) for some minimum period of time. These laws are having a benefit.
Are they a deterrent or do they just stop re-offending?
The only thing we can really measure well so far is that it’s stopping people from reoffending. In Washington, the state we’ve been able to evaluate most carefully, we have promising results that fatal crashes are being reduced as well. These are fatal crashes during the night time when alcohol is most likely to be a factor.
The other area where there’s progress in the US is automated enforcement. There’s some backlash against it, but more communities in the US recognising that they don’t have the police resources to monitor signalised intersections and enforce traffic light and speed violations. As this increases, we’re seeing fewer crashes and fatal crashes at red-lights.
We also have more communities looking at building roundabouts. Old technology, I understand, but sometimes an old low-tech solution is good. They move traffic efficiently and they’re a lot safer than an intersection with signals.
We’ve also lost a few battles in the US. With Michigan we lost yet another state in terms of mandatory helmet use. Michigan just rescinded this so now fewer than half the states have mandatory motorcycle helmet use. Given that the number of riders is increasing, this really is not the right direction.
We also still have a lot of states without primary belt use laws, where not belting up is a secondary offence that you cannot be stopped for. There’s even one state that has no belt use law. That’s New Hampshire. Their state motto is Live Free or Die. I guess they really mean it.
Are there any lessons learned?
Improving the safety of motor vehicle travel so people can have mobility without injuries is a long-term effort. We already have a strategy for this: research how the injuries happen, educate people about the problem, research solutions, then implement them and get people to support them. That’s the strategy we’ve been following for 30 years and I don’t think it changes.
Despite the improved safety, we still have 30,000 people being killed on our highways. In every single case, it’s a death or injury that didn’t have to happen. We’re looking for effective ways to intervene. It’s not magic; it’s just physics and biology.
In Europe and other markets there seems to be some polarisation of safety. The premium cars are adding a lot of technologies to improve safety, while some manufacturers are removing safety content, consciously providing something that complies with the legal standards but goes not much further. Are you seeing this in the US?
So far we don’t have that happening here. We only started releasing results last summer for our small overlap test and we’ve already got automakers striving to improve performance.
We complain about how slow our regulatory process is here. I think that if it were documented that automakers were consistently failing to pay attention to our tests – these are tests not required by the Federal Government right now – then our regulatory machinery would make that happen. They could soon show the positive benefit-cost ratio that they have to meet.
If you look around the world at what were previously thought of as secondary markets, people are increasingly unwilling to accept less than state-of-the-art safety in the products they buy. The progress in Latin NCAP’s results in their third round of tests is already much better than in the first round.
I can see why there might be some concern in Europe, but I think this may be a brief experiment until people realise that, ultimately, folks want to buy safe vehicles.
The new small overlap test accounts for around 25% of the frontal crash deaths in the US, but if the car makers respond really quickly and it ceases to be such a severe test for cars, would you then phase the test out?
We would phase it out if we could get the Federal Government to put it into regulation. But we are convinced that the automakers are responding because they know there will be a test. If we stop testing, there could be some backslide.
You think somebody would say, hey, we could save some metal on this structure here…?
I think the biggest problem would be getting to the end of designing a new product and it doesn’t do quite as well as it should do in this old test. Do you spend extra dollars to go back and see what went wrong or do you just release it as is?
If there’s no test there’s going to be a lot of pressure to release it as is. I think that’s what happens. I don’t think automakers start out saying we’re going to design an unsafe car because we can save money. It’s just that if nobody’s looking, you’re probably going to let it slide.
Okay, so how will your test protocols develop in the next few years?
We don’t have anything in the works but we are taking a look at side impact crashes now. We’re doing the same thing we did for frontals. We know that this test has been effective in the real world, but there are still people being seriously injured and killed in side impacts. We’ll see if there’s some upgrade that could address this.
First we need to understand what it is we’re missing. I’m not saying we will change, I’m just saying we’re looking at it. There are issues concerning the far-side occupants and this is something the current tests don’t address. We need to see if it’s a big part of the problem, if there’s a strategy for addressing it and a test to check the strategy.
What will be the big challenge for the next decade?
We’re going to focus a lot of attention over the next decade on researching advanced driver assistance systems. These go from collision warning systems to autonomous vehicles. We need to understand not just how systems behave on the test track but also how they behave on the road with real drivers.
We see a benefit for forward collision warning systems. With autonomous braking we see a bigger benefit. And for adaptive headlamps we’re seeing a benefit. But for lane departure warning we’re not seeing a benefit. We need to understand why this is happening because on the track and logically they all make some sense. We need some research to understand how drivers are using, or, in the case of lane departure warning, perhaps not using the systems.
If a car maker were planning a new platform for 2023, are there things that they should definitely put on the list of requirements?
Forward collision prevention systems clearly help. We will try to amend our rating system to reward the manufacturers who provide them. When we fully understand which technologies are effective, we’ll add them to our evaluations. We’ll be encouraging consumers to buy vehicles that not only protect you if you get in a crash but also that protect you by helping you avoid the crash in the first place.
The only way to make progress in this field is to keep coming back to the real world data. You have to ask how people are being injured and whether you can document a strategy that would fix those injuries and then if we have an adequate test.