State of the art in multiple response surface methodology

Abstract

The research begins with a brief history of multiple response surface methodology, including a description of the available multiple response surface methodology (MRSM) approaches. To facilitate the appropriate application of the available multiple response surface methodology approaches, the research includes the development of a structure for categorizing and evaluating multiple response surface methodologies (MRSM). Three schemes for categorizing multiple response optimization methodologies are developed: (1) type of approach (MRSM1 and MRSM2); (2) number and type of response variables (two means, more than two means, mean and variance of a single quality characteristic, and means and variances of multiple quality characteristics); and (3) preference elicitation (implicit and explicit). Each multiple response surface methodology approach is categorized according to these schemes. The application of multiple response surface methodology in product development is described, including the use of simulation with multiple response surface methodology approaches. Once underlying probability distributions are determined, simulation can be used to generate data ordinarily obtained using the experiments conducted in a laboratory, thereby producing a considerable cost savings in product development.