Auto engineers are taking advantage of expert systems to collect and reuse design experience, software to stimulate their thinking, and analysis to cause “accidents” every day.
Every time auto companies identify software that cuts a day off the product schedule, they stand to pocket $1 million. Conventional software, such as CAD and NC, is already in place. The biggest benefit will come from niche software — stuff that might guide engineers to discover new ways of solving problems and produce a spontaneous, “I never thought of that.”
The benefit of a wider range of software technology is already showing up in the price tags for the 1999 auto lineup. Only a few have been bumped up. Most car prices have held steady at no increase and a few have even been reduced.
How to put Einstein on your design staff
One unusual type of software finding its way into the auto industry stimulates the gray matter so users can find new ways around knotty problems. Such technology is helping several groups predict technology trends. One application is designing a better automatic restraining system.
Software that improves problem solving has been available in one form or another for several years. Early versions concentrated on general psychological methods. Recent versions have a strong scientific foundation and focus on tackling engineering problems.
One version, TechOptimizer from Invention Machine Inc., Cambridge, Mass., includes several sections that help state a problem with enough detail that a broader and hopefully more unusual solution surfaces. “Most products follow an S-shaped curve during development,” says James Kowalick, president of Renaissance Leadership Institute and automotive consultant. “For instance, products are invented, developed into a variety of forms to some technological level, and then mature with little further development until another invention makes them obsolete.”
For example, early versions of the carburetor were crude devices that poorly atomized gasoline. But they were refined over the course of the century and have been replaced with fuel injection systems. “A few forward-thinking automotive suppliers are using the problem-solving software to forecast technology based on the S-curve. So for fuel systems, instead of just increasing the injection pressure that might solve a short term problem, they are looking at other ways to push gas into cylinders.
You can see the S-curve at work at companies trying to develop hybrid gas-electric cars and all-electric versions. Kowalick says Ford Motor Co. thinks the problem-solving method is powerful enough to teach to over 800 of its people so far under the name SIT for Structured Inventive Technology.
The software also provides a new way of looking at airbags. The original version of the airbag inflates so fast that small people and children are hurt. A conventional solution to airbag injuries would be to simply reduce the inflation speed of the device.
“But slowing the airbag inflation rate reduces its collision protection in high-speed accidents,” says Kowalick. “At highway speeds, the chances of the airbag working to save your life are reduced because it’s not fully inflated when you hit it.” Reducing the airbag inflation speed solves one problem but creates another.
“A nonintuitive way to solve the conflict is to make the airbag inflate faster — even double the speed,” says Kowalick. “Now we’re not focusing on injuries caused by the airbag but on the main reason we have airbags, which is to protect people in severe accidents.”
But what if a small person is too close to such a bag when it inflates? The new problem associated with higher-speed bags is how to disable the device. “It’s a different problem with several solution avenues,” he adds. For example, a sharp device or laser could pierce it at a particular instant. This allows for always keeping the bag in a ready condition but puncturing it should it go off under circumstances adverse to an occupant.
This is less complicated than the so-called smart bags that several companies are working on. These will use many sensors to decide what if anything is in a seat (a person, dog, or groceries) and whether it’s in proper position. These more-complex systems are characterized by greater cost and lower reliability.
Kowalick says the proposed higher-speed system will always inflate and needs only a few sensors to decide on a person’s proximity. The simpler system is less than four years from production.