Optimization in manufacturing through experimentation and logistic regression models

Abstract

In this article, three optimization schemes are proposed for a response of sample non-conforming fraction type in a manufacturing process. The design of experiments is a statistical technique that is applied to improve the quality of the final product. The use of logistic regression models is recommended when analyzing the sample non-conforming fraction as a quality parameter. From the results of the experiment a mathematical model is constructed in order to optimize the process parameters. The proposed methodology was applied to the vulcanization process in the manufacture of rubber soles. A comparative analysis of three optimization methodologies is performed to evaluate the best result. The levels obtained in two procedures were the same and in the third one a less efficient solution was verified. The best solution was applied to the process with a significant improvement in the level of product quality.