Experimental Design Considerations for Accelerated Life Tests with Nonlinear Constraints and Censoring

摘要

As the electronic industry has adopted lead-free solder joint technology, the failure acceleration model for solder joint fatigue needs to be reinvestigated through accelerated life tests (ALT). However, traditional AL T designs currently employed by the industry fail to provide sufficient information for this purpose. This paper explores criteria for effectively planning AL T. It addresses the problem by hypothesizing a model form, deriving the model parameters using real data, and simulating ALT results for different experimental design plans. Relative comparisons are drawn between the influence of experimental design selection and some traditional reliability test plan variables such as sample size, censoring scheme, and sample allocation.