Genichi Taguchi (1924– 2012) was a Japanese engineer who worked on improving quality and reliability and established fundamental quality paradigms. Taguchi's methods and philosophies had a lasting impact on Japanese industry from the 1950s onwards, the effects of which meant that in the following decades many products from Japan could be offered more cheaply and also with better quality than European or North American products. The concept of the robust design or robust process originated with Taguchi. “Robust" means that the unavoidable disturbance variables – which Taguchi called noise factors– have as little impact as possible on the characteristics of the product orprocess. Likewise, a robust product is understood to be one on whose functions and properties the use conditions have as little impact as possible.
Taguchi developed his pioneering methods in the late 1940s and helped them become popular and widely used in Japanese industry. Many of his methods have been further developed inelectronic data processing, and many software systems offer the Taguchi method in a user-friendly way, meaning that nowadays the applications can be easily used without higher mathematical knowledge.
Methodology of design of experiment according to Taguchi
Today, Taguchi's methodology is classified as a special method within statistical Design of Experiment (DoE). The Taguchi method always focuses on minimizing the scatter involved in processes to make the product or process less susceptible to disturbance variables, i.e. to make it more robust.
Benefits and objectives
The core content of the two-day seminar is learning about the Taguchi methodology. The benefit for the participants or the users of the methodology is that after successful execution of the experiment it becomes clear where the scatter processes originate and with which parameter set of the controlled variables (according to Taguchi: control factors) the lowest possible scatter can be achieved. Processes with as little scatter as possible naturally lead to a higher and more stable yield. This significantly, sustainably and permanently reduces the costs for quality assurance, reworking and rejects for the manufacturer.
Taguchi’s methods are explained in the methodology environment of statistical design of experiment. Using practical examples, the methods based on Taguchi’s thinking are explained and discussed with participants. Practice examples complement the teaching.In the seminar, more advanced design of experiment methods are also discussed and practiced based on practical examples, which could only be done inadequately or rather awkwardly with the original Taguchi method.