Technology

Tech Innovations for Tough Times


How GE and others are using alternative techniques such as TRIZ to make R&D more efficient

For General Electric (GE), innovation isn't just about developing better technologies; it's also about not wasting cash on the wrong ones. That focus on saving money is even more important now that the economy has gone into a tailspin, dragging down GE's earnings and share price. The company's researchers have a surprising tool that could help: a method pioneered back in Stalin's Soviet Union.

It's called TRIZ, a Russian acronym for the phrase "the theory of solving inventor's problems." The core ideas were dreamed up by engineer and science fiction writer Genrich Altshuller, whose critique of the Soviet Union's record on invention in the late 1940s landed him in the gulag. There, he learned from imprisoned scientists and, when he was released, put together a step-by-step innovation method for people who aren't born with the gifts of Edison or Einstein. Since then, his theory has evolved into an elaborate system for analyzing problems and generating solutions. In contrast to brainstorming, TRIZ uses deep analysis of possibilities based on science and math algorithms.

These days, TRIZ is coming on strong at corporations hungry for new ways to improve innovation and productivity beyond what they've already achieved with the widely adopted Six Sigma and Lean techniques. In addition to GE, TRIZ fans include Intel (INTC), Samsung, and Procter & Gamble (PG), as well as smaller companies like FuelCell Energy (FCEL), a Danbury (Conn.) leader in power-generation fuel cells. The company employed TRIZ to evaluate the expensive flanges it uses to join pipes in its generators. After weighing the component costs, effectiveness, and complexity of assembly, FuelCell switched to a new clamping technique that will slash costs by 50%.

Rigorous Review Up Front

GE similarly uses TRIZ at the front end of the innovation process. Small project teams bring problems with them to TRIZ training sessions and use the method to evaluate them, come up with potential solutions, do risk analysis of the alternatives, and then devise conceptual designs that can be tested. By subjecting their ideas to this kind of rigorous review up front, GE scientists hope to avoid costly mistakes down the line. "Conceptual design is a vital step," says Michael Idelchik, vice-president for advanced technology at GE Global Research. "If you start with an elephant, no matter how you optimize it later, you'll never have a giraffe."

A number of specialty consulting firms have built practices around TRIZ. GE is working with GEN3 Partners. Though based in Boston, GEN3's 150 employees mostly work from Russia. Others are Russian émigrés, including President Semyon Kogan and Simon Litvin, GEN3's chief science officer, childhood friends who grew up in what was then Leningrad to become scientists during the Soviet era.

GEN3 has added some of its own twists to TRIZ. It taps into a global network of 8,000 science experts and finds solutions for problems in one industry by locating applicable technology in another. "Don't invent if you don't have to," says Kogan. "Go and find solutions that have been used and apply them to your problem, and you'll have a higher probability of success." Under GEN3's tutelage, GE found technology in the radar industry for use in its next-generation MRI machines.

It's difficult to predict how big a deal TRIZ will become. University of Michigan professor Jeffrey Liker, an expert in the Lean manufacturing discipline, calls it a niche tool. But GE's Idelchik says it could provide big paybacks. So far, 382 GE employees in 70 teams have completed the training. Supervisors have green-lighted 90% of their conceptual designs, and product development is now under way. Not bad for an obscure theory born in Stalinist times.

Hamm is a senior writer for BusinessWeek in New York and author of the Globespotting blog.

Toyota's Hydrogen Man
LIMITED-TIME OFFER SUBSCRIBE NOW

Sponsored Links

Buy a link now!

 
blog comments powered by Disqus