Monitoring process for attributes with quality deterioration and diagnosis errors

• Published in: Applied stochastic models in business and industry Vol. 23; no. 4; pp. 339 - 358
• Main Authors: Trindade, Anderson Laécio Galindo, Ho, Linda Lee, da Costa Quinino, Roberto
• Format: Journal Article
• Language: English
• Published: Chichester, UK John Wiley & Sons, Ltd 01.07.2007
• Subjects: economic model; Markov chain; misclassification errors; on-line attribute monitoring; quality control; quality deterioration
• ISSN: 1524-1904; 1526-4025
• DOI: 10.1002/asmb.675

The aim of this paper is to present an online economical quality‐control procedure for attributes in a process subject to quality deterioration after random shift and misclassification errors during inspections. The process starts in control (State I) and, in a random time, it shifts to out of control (State II). Once at State II, the non‐conforming fraction increases according to a non‐decreasing function ψ(z), where z is the number of items produced after a shift. The monitoring procedure consists of inspecting a single item at every m produced items, which is examined r times independently to decide its condition. Once an inspected item is declared non‐conforming, the process is stopped and adjusted. A direct search technique is used to find the optimum parameters which minimize the expected cost function. The proposed model is illustrated by a numerical example. Copyright © 2007 John Wiley & Sons, Ltd.