Fundamentals of Design of Experiments and Optimization: Experimental Designs in Response Surface Methodology

Abstract

Designing an experiment to fit a response surface model typically involves selecting a design among several candidates. There are often many competing criteria that could be considered to properly choose the final design. This chapter presents a description of the most reported designs involved in the data collection process. The basic principles and characteristics of the designs, highlighting their strengths and weaknesses, are described in detail and compared in terms of their characteristics and efficiency. The information is divided into process designs, mixture designs, and designs with the combination of process variables with mixture variables. Additionally, the use of constraints in the delimitation of experimental spaces in both process and mixture designs is also introduced and discussed. Moreover, a brief description of non-randomized designs is presented.