National Transportation Week II
You introduce the idea of “From Road to Lab to Math.” What do you mean by this?
The idea is that our understanding of how to design vehicles best to protect us in crashes has evolved over time. In the industry, engineers first began by examining wrecks that occurred on the road (real crashes) and tried to see what features of the vehicle helped or hurt people. Then, testing moved into the laboratory where controlled experiments could happen. Now, testing is moving primarily into computer simulation where you don't actually build physical vehicles. Instead we run virtual crash tests. Basically, road to lab to math is a progression of testing capabilities to make vehicles safer and to do it more cheaply.
What are some of the benefits to computer-based simulation technologies?
The biggest benefit to using computer simulation for the consumer is that engineers can do many many more design iterations. That is, they can test many more vehicle designs than they could before to see which ones best protect occupants in collisions. In an ideal world (we're not there yet), engineers would be able to optimize the design of the vehicle so it is the safest possible by triangulating between multiple design solutions. In the world of physical testing, it takes so long to build real vehicles to test in the lab and it costs so much that the number of tests engineers can run is much smaller and, consequently, they don't have as much data to arrive at an optimal design.
What are some of the drawbacks?
The major drawbacks is that simulations are not always entirely accurate. They can have problems with them and their dynamics don't always match the results of physical tests. This is less and less of a problem all the time. The larger problem is that within companies there is still a distrust often times of computer simulations and so managers encourage more physical tests, which then negates some of the benefits of increased use of simulations.
Why have companies begun to turn away from math-based simulations?
They have not. In fact, they embrace them more and more all the time primarily because they are more cost efficient. It can cost as much as $700k to build and crash a car into a wall to test its safety. The cost is just a few dollars to do the same in a simulation and you can do the simulations much faster. So overall, (even despite some people's distrust of them) simulations are becoming more and more common.
In Car Crashes without Cars, you say that even during this economic downturn, car companies still hire computer-based simulation engineers. Why are these engineers so important to automakers?
Primarily because of cost. At the automaker I studied there were massive layoffs of people who conducted tests of physical vehicles at the proving grounds, but virtually no layoffs (and increased hiring, in fact) of engineers who did simulations. This was because the companies are trying to move away from hardware testing because its so expensive to test and to maintain the facilities. Simulations are a way to keep costs down and get to market faster.
The theme of this year’s National Transportation Week is “The World is Right Next Door” and celebrates the role transportation has played in connecting the world. Would you say that multinational car companies have helped foster global collaboration?
Yes, for sure. Certainly by making mobility easier for people. But auto companies also do a lot of offshoring of design and engineering work to captive offshore centers. Diane Bailey and I discuss this phenomenon in our book due out from MIT press in 2014, which is tentatively titled, Technology Choice: Why Occupations Differ in their Embrace of New Technologies. They also source parts from all over the world. So there is a big global footprint in the auto community. Also, in terms of vehicle safety, sometimes designing a vehicle for safety requirements in one country and selling that vehicle in another county helps lawmakes in those other countries increase the stringentness (is that a word?) of their safety laws because the vehicles are capable of achieving them. So, there can be more uniformity across countries in terms of what we expect from our cars.
Or, if you prefer to use an RSS reader, you can subscribe to the Blog RSS feed.RSS
- art (5)
- cognitive science (2)
- computer science (4)
- current affairs (3)
- design (5)
- digital humanities (1)
- economics (3)
- education (1)
- engineering (2)
- environmental studies and nature (6)
- games (2)
- history (1)
- humanities (1)
- information science (4)
- journals (1)
- literature & poetry (1)
- math (1)
- MIT (1)
- neuroscience (2)
- new media (1)
- philosophy (2)
- robotics (3)
- science (2)
- technology (11)
- web / tech (2)
- women (1)
Blogs We Like
Books, news, and ideas from MIT Press
The MIT PressLog is the official blog of MIT Press. Founded in 2005, the Log chronicles news about MIT Press authors and books. The MIT PressLog also serves as forum for our authors to discuss issues related to their books and scholarship. Views expressed by guest contributors to the blog do not necessarily represent those of MIT Press.