Collaborative Research: Design of Equivalent Accelerated Life Testing Plans Involving Single or Multiple Stresses

新产品推出的激增，加上从产品设计到制造的时间大幅缩短，以及客户对高可靠性和更长的保修期的期望不断提高，促使行业缩短了产品测试时间。在许多情况下，在实际生产之前，产品原型的测试可能需要数月甚至数年的时间。在许多情况下，将产品样品置于极端条件下的加速寿命测试可能是评估产品是否满足预期长期可靠性要求的唯一可行方法。因此，来自此类测试的可靠性估计的准确性对有关系统配置、保证和预防性维护计划的后续决策具有深远的影响。测试执行不当可能导致产品发布延迟、整个产品终止或产品实际使用中的灾难性故障。在各种工业应用中，在进行测试时，应力载荷可以有许多选择。每种应力加载都有其优缺点。这就提出了许多关于涉及不同应力载荷的各种加速测试计划之间的等效性的实际问题。当涉及多重压力时，问题变得更加困难。为了克服这些挑战，我们建议调查和贡献基础和理论模型，并提供一些统计工具来促进实际的可靠性测试。建立规划各种应力载荷下加速试验的新框架，特别是复杂应力和多重应力条件下的加速试验。如果获得成功，研究成果将填补行业所需的可靠性评估的重大空白。特别是，测试结果将使工业界能够设计高效和经济的测试计划，而测试结果将提供准确的可靠性估计。所开发的工具将促进和进一步推动加速寿命试验在现代工业中的实施。

The surge in the introduction of new products coupled with the significant reduction in time from product design to manufacturing, as well as the ever increasing customer's expectation for high reliability and longer warranties have prompted industry to shorten its product test duration. In many cases, the testing of the products prototypes can take months and even years before actual production takes place. In many situations, accelerated life testing which subjects the product samples to extreme conditions might be the only viable approach to assess whether the product meets the expected long term reliability requirements. As a result, the accuracy of reliability estimates from such testing has a profound effect on the subsequent decisions regarding system configuration, warranties and preventive maintenance schedules. Inappropriate implementation of the test might cause delays in product release, termination of the entire product or catastrophic failures in the actual use of the product. In a variety of industrial applications, there could be many choices in stress loadings when conducting testing. Each stress loading has some advantages and drawbacks. This has raised many practical questions regarding the equivalence among various accelerating test plans involving different stress loadings. The problem becomes more difficult when multiple stresses are involved. To overcome these challenges, we propose to investigate and contribute fundamental and theoretical models and provide several statistical tools to facilitate practical reliability tests. New framework for planning accelerated testing under various stress loadings, especially under complex stress and multiple stresses conditions, will be established. If successful, the research results will fill a major gap in reliability assessment needed by industry. Especially, the results will enable industry to design efficient and economical test plans, yet the results of the test will provide accurate reliability estimates. The tools to be developed will facilitate and further promote the implementation of accelerated life testing in modern industry.