Planejamento econômico de controle on-line por atributo

Abstract

The economical design for online process control by attributes, proposed by Taguchi, Elsayed and Hsiang (1989), consists of inspecting a single item at every m items produced. It is assumed that the process starts operating in-control. If the inspected item is non- conforming, the process is designated as out-of-control and production is stopped for an adjustment; otherwise, the production goes on. Given this premise, several authors expanded Taguchi’s model incorporating additional features such as: use of probability distributions to describe the process shift, presence of misclassification errors on inspections, incorporation of variable interval sampling and non-unitary samples. This study presents a review about the economical design of online process control by attributes, discusses the mechanisms of genetic algorithms and its applications in statistical process control, and finally, performs simulations and analysis comparing the different models presented.