A Matter of Survival Simulation and statistical methods help manufacturers achieve high product quality now essential in competing on the world market.

Products designed using sound principles often fail once they are built and in use. That’s one of the frustrating puzzles of engineering: defective products based on designs that passed quality checks, engineering analysis and prototype testing with flying colors. In many cases, such failures are caused by the unforeseen interaction of multiple variables in production, materials, shipping and customer use — and manufacturers pay a steep price for these failures. The key to evaluating these kinds of interactions is an approach called Design of Experiments (DOE), in which numerous random analyses are run on different combinations of changing variables. Probabilistic and statistical methods compare all the different results and study the sensitivity of product behavior to these variations. The tools are at the heart of Design for Six Sigma (DFSS) methods in arriving at optimal near-defect-free “robust” designs that work properly — even in the face of wide variations of product parameters. For years, teams of statisticians, analysts, designers and experts have had to spend months plowing through thousands of simulations and mountains of data for such studies. Consequently, the DOE approaches were used