Reliability Improvement Experiments With Degradation Data

Abstract

Design of experiments is a useful tool for improving the quality & reliability of products. This article develops an integrated methodology for quality & reliability improvement when degradation data are available as the response in the experiments. The noise factors affecting the product are classified into two groups which led to a Brownian motion model for the degradation characteristic. A simple optimization procedure for finding the best control factor setting is developed using an integrated loss function. The methodology is illustrated with an application to a window wiper switch experiment.