Taguchi Engineering in Robust Design Optimization

When it comes to the field of design of experiments methodology, Genichi Taguchi, a Japanese engineer and author is one of the best known names around. Having published his first book on experimental design in 1958, Taguchi’s methods of fractional factorial design have forever earned him a place in history when it comes to design of experiments. Taguchi methods are statistical methods, which were originally developed to improve the quality of manufactured goods. They have more recently also been applied to engineering and biotechnology. Taguchi method, also known robust design, greatly improves engineering productivity. Robust design optimization focuses on studying the impact of variations and on how to make a process or product more stable when having to account for or deal with variations over which there is little or no control. It therefore works to improve the function of the product or process, and facilitating flexible designs and concurrent engineering. This makes it an extremely powerful technique to reduce cost, improve quality and speed up development time. Taguchi engineering categories variables into two distinct types. First, there are control factors, which are variables that can practically and affordably be controlled. These would be factors such as dimensions and material parameters. Next, there are noise factors, which in contrast to control factors are difficult or expensive to control. Noise factors might be able to be controlled in an experiment, for example temperature or environmental exposure, but they can’t be guaranteed to be controlled once the product or process is taken outside of the experimental landscape. By considering these two factors that Taguchi method works to produce the most robust design by determining a combination of control factor settings that will produce a product with the maximum ‘robustness’ to accommodate the expected variation in the noise factors. By consciously considering the noise factors and the cost of failure in the field, the Robust Design method helps ensure that the very best product is produced. Taguchi engineering, for example, would be useful when developing operating systems for planes that will be subjected to a variety of different environmental variations, humidity fluctuations and the like. Several advanced software solutions, such as Optimus software, offer Taguchi engineering methods as part of their solution methods. Based on orthogonal arrays, Taguchi design of experiments works to provide a reduced variance for the experiments with an optimum setting of the control factors. The resulting design robustness is enabled through a rich and complete set of Taguchi functions.