A variables-type multiple-dependent-state sampling plan based on the lifetime performance index under a Weibull distribution

• Published in: Annals of operations research Vol. 311; no. 1; pp. 381 - 399
• Main Authors: Wang, To-Cheng, Wu, Chien-Wei, Shu, Ming-Hung
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.04.2022; Springer; Springer Nature B.V
• Subjects: Acceptance tests; Advanced manufacturing technologies; Business and Management; Combinatorics; Dependent variables; Design; Design parameters; Failure; Industrial engineering; Lifetime; Manufacturing; Methods; Operations research; Operations Research/Decision Theory; Optimization models; Performance indices; Product design; Product quality; Product testing; S.I. : Reliability Modeling with Applications Based on Big Data; Sampling; Suppliers; Theory of Computation; Variables; Verification; Weibull distribution; Taiwan; OC curve; Lifetime performance index; Weibull distribution; Acceptance sampling plan; Type-II right censoring; Multiple-dependent-state sampling plans
• DOI: 10.1007/s10479-020-03655-z
• ISSN: 0254-5330

Verification and validation of product quality highly affects the success of product design and manufacturing processes as well as the long-term collaboration between buyers and suppliers. Lifetime testing for acceptance is the main bottleneck in verification and validation and often requires considerable time and expense, especially in the current high-product-yield era. To establish a cost-effective and risk-controllable life testing sampling scheme, we propose a variables-type multiple-dependent-state (MDS) sampling plan using the lifetime performance index under a Weibull distribution with Type-II right censoring. The design parameters of the lifetime-capable MDS sampling plan are formulated as optimization models to minimize the required number of testing failures subject to nonlinear constraints for the desired lifetime capability levels and allowable risks regulated by the supplier and buyer. Compared with variable single sampling plans, the plan proposed herein is more cost-effective sampling and has greater discriminatory power. The applicability of the proposed plan is demonstrated through an example.