Optimization by Design of Experiment techniques

Abstract

Design of Experiments (DOE) is a statistical technique for quickly optimizing performance of systems with known input variables. It starts with a screening experimental design test plan involving all of the known factors that are suspected to affect the system's performance (or output). When the number of input variables or test factors is large, the primary experimental objective is to pare this number down into a manageable few. This is usually followed by another designed experiment design or test plan with the objective of optimizing the system's performance. The most common initial and final optimization designs of experiment are called the Screening Design and the Response Surface Method (RSM). This paper will present some examples in the use of these designs.