Repetitive Testing Strategies When the Testing Process Is Imperfect

• Published in: Management science Vol. 44; no. 10; pp. 1367 - 1378
• Main Authors: Ding, Jie, Greenberg, Betsy S, Matsuo, Hirofumi
• Format: Journal Article
• Language: English
• Published: Linthicum, MD INFORMS 01.10.1998; Institute for Operations Research and the Management Sciences
• Series: Management Science
• Subjects: Accuracy; Applied sciences; Computer; Computer industry; Computer/Electronic Industry; Cost efficiency; Costs; Electronic Industry; Electronics; Electronics industry; Exact sciences and technology; False negative errors; Inspection; Laboratory equipment; Manufacturing; Mathematical vectors; Minimization of cost; Optimal Testing Strategies; Optimization; Probability; Product testing; Production technology; Quality control; Random access memory; Semiconductor Manufacturing; Semiconductors; Sequencing; Studies; Testing, measurement, noise and reliability; Unit costs; Variable costs; Random access memory(RAM); Optimal strategy; Semiconductor materials; Electronics industry; Inspection; Computer; Delay time; Optimal test; Repetition
• ISSN: 0025-1909; 1526-5501
• DOI: 10.1287/mnsc.44.10.1367

This paper considers the problem of test design and implementation when testing is imperfect. Items that are classified as conforming may be nonconforming, resulting in a poor outgoing quality level. Items that are classified as nonconforming may be conforming, resulting in excessive scrapping of conforming items. The failed items are commonly retested to reduce the scrapping problem. Alternatively, the accepted items may be retested to improve outgoing quality. In this paper, we examine the question of whether it is better to repetitively test rejected items, or to repetitively test accepted items. We also examine the relationship between the two testing policies, testing equipment accuracy and capacity, incoming quality, and outgoing quality requirements.