الفهرس 
Title : Optimal design of accelerated censored test plan using alphabetical optimality criteria and mathematical programming
AbstractOnly 14 pages are availabe for public view
 |  | from | 122 |  |  |
| | | from | 122 | | |
Abstract
Accelerated life testing (ALT) is a technical strategy used for assessment of highly reliable products depending on actual failure of tested units. While accelerated degradation testing (ADT) is another important type of accelerated testing that depends on performing degradation other than product failure. Planning accelerated tests is an important aspect that has been extensively studied recently. Statistical precision for estimates is an important factor considered by researchers in planning accelerated tests. While experimental cost is a major consideration for manufacturers and has to be considered as an important objective. Accordingly, this study aims at achieving the compromise between precision and cost by proposing a multi-objective goal programming. For ALT, the study designs an optimum multi-objective ramp stress accelerated life test under progressive type I censoring. Model parameters are estimated using maximum likelihood estimation (MLE) method. Problem of determining number of removals is addressed using D-optimality criterion. A goal programming model is built as an alternative approach for designing optimal test plan. For illustration, a comparative study is conducted between the two approaches. Designing an optimal step stress accelerated degradation test (SSADT) is also considered. A goal programming model that minimizes weighted sum of asymptotic variance of mean time to failure and the experimental cost is presented. Some design variables as sample size, measurement frequencies and number of measurements under different stress levels are optimized under the constraint of increasing number of measurement with higher stress levels. A comparative study is performed between proposed model and previous study. The sensitivity of the SSADT plan is also studied