Designing an accelerated degradation test plan considering the gamma degradation process with multi-stress factors and interaction effects. Issue 5 (2nd September 2020)

• Record Type: Journal Article
• Authors: Limon, Shah; Rezaei, Esmaeil; Yadav, Om Prakash
• Is Part Of: Quality technology & quantitative management. Volume 17:Issue 5(2020)
• Page Start: 544
• Page End: 560
• Publication Date: 2020-09-02
• Subjects: Accelerated degradation test -- gamma process -- multi-stress factors -- interaction effects -- monotonic degradation -- asymptotic variance; Quality control -- Periodicals; Quality control -- Statistical methods -- Periodicals; Industrial management -- Periodicals; Industrial management; Management -- Research -- Methodology -- Periodicals; Qualitative research -- Periodicals; Management; Quality control; Quality control -- Statistical methods; Periodicals; 658.00721
• DOI: 10.1080/16843703.2019.1696010
• Languages: English
• ISSNs: 1684-3703

Abstract

ABSTRACT: The effectiveness of accelerated degradation test and subsequent reliability assessment depends on the efficient test plan. Designing an accelerated test plan requires determining the optimal values of several decision variables. The problem of designing an accelerated degradation test becomes even more challenging when multiple stress factors and their interaction effects are considered. In this work, an optimal accelerated degradation test plan is presented considering the stochastic gamma process with constant stress loading. Both gamma parameters are treated as stress-dependent variables to capture the impact of stress levels. An optimization model is formulated considering multiple stress factors and their interaction effects for offering a more realistic test plan. The optimization criteria are set to minimizing the asymptotic variance of the maximum likelihood estimator of the lifetime at normal operating conditions within the total experimental cost constraints. The decision variables comprised in the optimization model are measurement time intervals, the frequency of measurements, stress levels, and sample allocation at each stress level. The sensitivity analysis is performed to investigate the model's robustness against the budget constraints and pre-estimates of model parameters. To demonstrate the proposed methodology, a case example of carbon film-resistors' performance data is considered for designing the degradation test plan.