Priya Prasad got started in biomechanical research by studying the effects on the body from ejecting out of a jet fighter.
Prasad, a senior technical fellow at Ford Motor Co., specializes in automotive safety. He is spearheading an effort to accurately model the human body with computers for simulated crash testing. Prasad has also been the company's point man on issues such as sport-utility safety and vehicle compatibility.
He spoke last month in his Dearborn, Mich., office with Staff Reporter Aaron Robinson. An edited version of the interview follows.
Why does a car company want to model the human body with computers?
If you have a model of the human body, you can scale it up or down. You wouldn't need to run all these discreet sizes of dummies in crash tests. It could be simulated with computers.
The material is the big issue: We know steel; we know aluminum; but we don't know how human tissues react in all situations. We started with the head and neck because head injuries are very serious contributors to death and disability.
Where it will really help out is in post-crash care. Many people have injuries that can't be seen. This can tell doctors that these are the organs you need to concentrate on.
To what point do you see computer simulations advancing in the next 10 years? And do you think that Ford will ever be confident to deliver a car for barrier testing that has been completely computer simulated?
I have a very high confidence level on the prediction technology. It doesn't have that much experience because they're used to crashing cars. I think that in 10 years we'll be able to convince (the government) that computer technology is advanced enough for computer-aided compliance. It doesn't mean that all testing will stop, but they should be checking out our cars through models.
What has been the most important safety advancement in the last 30 years?
The conventional three-point seat belt. It's effective in all crash modes, including the front, side, and of course in rollovers.
I don't think that a giant leap like that is going to happen with restraint systems. There will be some evolutions in restraint systems, but the real gain in safety now will have to come to pre-crash sensing and post-crash notification.
Restraint systems down the road might become adaptive restraint systems. So it will adapt for speed of the crash and the size, age, and weight of the occupants. Injuries are very much a function of speed as well as age and gender. So 10 years down the road, maybe we'll have a system that can adapt to the full circumstances of the crash and the occupant.
Where do cost and benefit intersect concerning safety features?
Any safety device that is too expensive, in my mind, is not a safety device if you can't put it out. Obviously, we cannot make safety unaffordable.
The challenge in safety is to provide affordable safety. You can have all sorts of gadgets, but if they are too expensive, nobody will be able to afford them and that hasn't done much for safety in that situation. But we don't do cost-benefit analysis as such.
What factors go into deciding what safety systems will be incorporated into a vehicle?
If we see the real world benefit - and 'real world benefit' means 'benefit to the majority of the population' - then we will look at reducing the cost of that technology and putting it in mass scale.
It's quite possible that the first application of the technology may be in a more expensive car or truck. However, at Ford we like to start cascading the application to all our products. If you look at the side airbags, it's going across the board now, but it started off in more expensive cars.
What evidence was there that side airbags were a beneficial feature?
The evidence that we had was essentially for head protection, not so much for chest protection. We could design the side of our cars to provide protection on side impact for the chest and the pelvis and also the abdomen, but there was nothing that would provide protection to the head. That's when we at Ford came up with the idea that any side airbag that can also protect the head is the one that will be most beneficial in the real world.
To prove it out, we originally started with math models. We ran these math models of cars running into poles or trucks running into side of cars. We found that head airbags will have tremendous lifesaving potential.
During the development phase, we actually ran some tests with combined head and chest airbags, and we found that, in full impact, there was substantial benefit to the head. With the airbag, probability of a head injury was minimal.
So the decision to continue applying the system was drawn from computer modeling?
Yes. And also we started the accident later. Quite a few side impacts with poles. When you look at the accidents that have pole impacts, head injuries are very high. And when you look at trucks running into sides of cars, head injuries in those situations are also very high. So that gave us the idea that we should go for head protection.
Now that you have a population of vehicles out there with this safety technology, are you getting real world feedback?
It always takes a few years to get the actual data. You need enough of this data to come up with statistical conclusions.
Several institutes have actually run tests with and without the head-chest airbags in pole impacts and truck impacts. All the tests are coming out real good.
The field data hasn't come in yet. Good, solid statistical data takes time to build up.
Do you know when you'll be able to make a determination on that?
It will be at least a couple of years. Ford has put out head and chest airbags. Other companies put out just the chest airbags. So to really collect the data on head-chest airbags will take time. We need people to get involved in accidents and then figure out how they're working. I personally believe that they'll work very well.
In light of problems with airbags, there's been some debate about whether safety technologies are doing more harm than good. Is that an issue?
That's true for any restraint system that's going to deploy during the crash because you are adding energy. The frontal airbag situation was somewhat strange in the sense that we were forced to actually put them out by law. Congress dictated that you must have front, outward airbags.
The second aspect of that airbag was the regulation, which said that you had to design it to protect a 50th percentile male dummy from mid-seat position against a rigid barrier at 30 mph. So just to meet the regulation, it required a certain amount of aggressivity of an airbag. And it's hard to imagine that you can have an airbag that is benign to children in the 30- to 40-pound range and also stop a 165-pound occupant.
If you could rewrite any one automotive regulation, what one would it be?
I would rewrite FMVSS 208, which is the airbag regulation. I would design a restraint system for belted occupants, even though I know 30 percent of the people don't wear their belts. There's a tremendous tradeoff that's going on to design for these unbelted people. We are trading off some injuries for even belted people.
I know that leg injuries have gone up for belted and unbelted people in all cars because we have had to design knee bolsters at high-force levels to meet the unbelted regulation. For the 30 percent of irresponsible people, we are taking that 70 percent and making them go through higher risk.
How could a car be designed to better protect an unbelted occupant?
For belted occupants, you can move the knee bolster forward so a lot of the hard contact with the knee bolsters that occurs now would not take place. These hard contacts happen because we are really designing for unbelted occupants. Injuries occur because there is contact and the contact has to be hard. But if the occupant is belted, we could move the knee bolster forward, away from the occupant, and make them softer. If you look at the data on leg injuries, they've gone up with the passive restraint regulations after they were phased in.
How would you write the regulation for a belted occupant?
I would write it around a belted occupant maybe at 35 mph instead of 30, because we believe we can provide protection up to 35 into a rigid barrier.
Assuming the standard remains the same, what is the solution to the increase in leg injuries?
There's a lot of research going on for that. We have established injury criteria for a test procedure, which is the offset test now. We are basically designing our cars to meet those injury criteria. It's something the U.S. doesn't have, by the way. It is a European rule.
Will that involve a passive system?
I don't think so. People always talk about inflatable car parts. We can get there without these devices. Quite a few of our cars have done that.
Does that mean you're opposed to technology-based solutions?
My approach is performance-based. We should have performance standards, but if there are situations where you cannot get to your performance without high technology - those are the situations where we go into technology.
Safety systems as a percentage of the price of the car are growing. Ultimately, what's going to happen to that trend?
Affordability is important. I don't know where it stops. Obviously there's a point where the technology will become unaffordable. Where that price point is, I really don't know.