Design process of spoke-type PM motor using six sigma process with tolerance design

Abstract

Six sigma process (SSP) is a process of providing the perfect products without any defects for their customers by considering both mean and variance of manufacturing imperfections. A few researchers have presented some papers on motor and/or drive considering performance variance as well as mean using the SSP. However, all of the presented papers cannot focus on the overall SSP, but mainly optimization step taking into account the probability of failure (POF). Therefore, there was lack of explanation for parts such as analysis of the effects of design factors (variables) on responses (objectives), how to minimize the number of optimized factors and how to optimize design factors considering POF by tolerance design (TD), and so on. This paper presents the full details of SSP from defining objectives to verifying design results with an example study of spoke-type PM motor design, which is applicable to motor design process including the above mentioned terms. SSP is composed of the five sequential steps i.e. Define-Measure-Analyze-Improve-Control (DMAIC) steps, while some analysis tools and statistical theories should be considered in each step. First of all, two important indexes i.e. Z-value and p-value along with the statistical analysis are explained in detail. Next, some useful tools for each step are explained e.g. Y-y mapping method in define step, Gage R&R in measure step, analysis of variance (ANOVA) in analyze step, response surface methodology (RSM) and tolerance design (TD) in improve step, and monte-carlo simulation (MSC) and process capability analysis (PCA) in control step, respectively. Finally, it is shown that the SSP is useful for motor design process to satisfy both the required performances and POF of spoke PM motor.