Quality Improvement in the Multi-response Problem by Using Clustering Characteristic

Most previous studies have only addressed a single-response problem. However, more than one correlated response frequently occurs in a manufacturing product. The multi-response problem has received limited attention. In the second part of this project, an approach based on the clustering analysis (CA) is studied to the optimization of the multi-response problem. In CA, the observations can be combined into groups or clusters such that each group or cluster is homogeneous or compact with respect to certain characteristics and each group is different from other groups with respect to the same characteristics. The optimum parameters’ settings for a multi-response problem can be determined by three criterions. Key-Words: clustering analysis, quality improvement, parameter optimization