Design and Conquer

The Joint Strike Fighter may be the most complicated manufacturing effort ever. How will they do it? With IT, and lots of it.

In 1994, the Pentagon issued the defense industry a challenge: design a jet fighter that satisfies a mass market (the U.S. Army, Navy, Marines and allies) but is also customizable (Want to take off from an aircraft carrier? No problem. Land vertically? Done). Last fall, Lockheed Martin of Fort Worth, TX, won the Joint Strike Fighter contracta victory estimated to be worth $200 billionnot only because of its design, but also because of its new economy promise: to employ state-of-the-art software to marry the fighter's design and manufacture processes, and connect hundreds of subcontractors around the world.

Lockheed Martin is working to fulfill that promise by paring down the number of software packages used to design and build the plane and tightly integrating the ones that remainto support the most complex and far-flung projects in military history. It selected one applicationMetaphase, from Plano, TX-based EDS, to manage its so-called product data: all of the information about the fighter's thousands of parts, its vendors, and the workflow of its production. Lockheed selected a second application for computer-aided design (CAD), the French firm Dassault System's CATIA. To link them, Lockheed Martin selected Lisle, IL-based T-Systems' software, CATIA-Metaphase Interface.

The firm has begun to connect key partnersthe first of thousands of engineers at hundreds of suppliersto its system; as a result, a design team in Fort Worth will be able to see what configurations are available for a pump manufactured in Taiwan; a subcontractor in London will be able to seein real timechanges to the configuration of a hydraulic system and revise their plans accordingly. "The old [archetype] was the designers throw the design over the wall to the manufacturers and say 'design this,'" says Tom Salva, vice president for manufacturing at T-Systems, a spinoff of Daimler Corporation and now a subsidiary of Deutsche Telekom. By linking design and development, Lockheed Martin hopes to cut the time it takes to get the wings on and reduce costs associated with design changesone of the key requirements for the multimillion dollar fighter, the first of which are scheduled for delivery in 2008.

"The product structure you have in the CAD system is also part of the product data management system," says Mike Ballard, a business systems analyst at Lockheed Martin. "[Linking the two] lets you drive the CAD system with the product data. CAD developers tend to do more and more integration when they cross those boundaries."

Further complicating matters are the various levels of access each partner has to the design, much of which is, understandably, classified. "A U.S. citizen working in Fort Worth looks at the data," Salva says. "You put that same employee on a plane and send him to British Aerospace, and he can only see a subset of what he could see in Fort Worth. And that's still more than his British counterpart, at the next computer, can see."

Story continues below

Linking design and development information from the start could help Lockheed Martin meet critical cost and performance deadlines, says aerospace and defense analyst Michael Burkett of AMR Research. But technology that's state-of-the-art at the beginning of a development cycle may be dated by the time the plane hits full production2010, in the case of the Joint Strike Fighterand a historical artifact when militaries are maintaining planes up to 40 years from now. "It's a very typical problem in the aerospace industry," Burkett says. "They've got funding to buy all this new software, but when it gets to maintenance mode, that well will be dry."

But Lockheed Martin engineer Glenn Hayhurst says his company is working hard to keep its software up-to-date, a challenge that will grow as Lockheed adds partners and ramps up production over the next decade, growing, eventually, into the largest defense program in historyand, Hayhurst and his colleagues hope, an historic model for large enterprises.