Out with the old

Building large, complex systems, such as military vehicles and aircraft, is a time-consuming and often wasteful and unpredictable process. The problem is that manufacturers rely on a method that has been in place since the dawn of the Industrial Revolution: design, prototype, test, correct and retest. The Defense Advanced Research Projects Agency (DARPA), however, wants a new amphibious vehicle for the U.S. military built in about one-fifth the time it would typically take—along with the corresponding cost reduction—while achieving the desired performance.

In an attempt to achieve that, the agency created FANG (Fast, Adaptive, Next-generation Ground vehicle) design challenges, with the first of three launched early this year. DARPA invited engineers and students to use the tools developed by universities and agencies working on the problem.

One of the schools is Oregon State University. OSU is developing a concept called “model-based design and verification,” where virtually all of the design, testing, error identification and revisions will be done on a computer up to the point of commercial production, explained Irem Y. Tumer, associate professor of mechanical engineering at OSU’s School of Mechanical, Industrial and Manufacturing Engineering.

“Most of our work looks at how component failures in a large system like an amphibious fighting vehicle will propagate through the system and then simulate what specific functionality will be lost,” Tumer said. “We want to know early on in the design stages, before we actually have detailed models, what functionality we should be paying attention to, what is safe and what is unsafe, and what types of reliability problems we will have.”

The researcher’s failure propagation analysis involves examining the architecture of a large system to determine how the components are arranged and connected and the types of material, energy and information that flow among them, Tumer added. Then it looks at the functionality of each component. This can uncover hidden component dependencies, which become important when a fault occurs because it helps identify what component failed.

“We’ve done a lot of work like this in the past with individual parts and small groups of components,” Tumer said. “Now we’re taking that complexity to the level of a finished and completed machine, sometimes with thousands of parts working together.”

In theory, a new machine, such as a vehicle, should work right the first time and perform exactly as the computer said it would using model-based design and verification. “We’ll see what works, identify and solve problems, make any desired changes and then go straight to commercial production,” Tumer said. “If this works, and we believe it will, it will revolutionize the way machines get built.”

For more information about OSU’s School of Mechanical, Industrial and Manufacturing Engineering, contact Tumer at (541) 737-6627 or visit mime.oregonstate.edu. For more information about DARPA’s FANG design challenges, visit www.vehicleforge.org.