Experimental design considerations for accelerated life tests with nonlinear constraints and censoring

Problem: 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 ALT designs currently employed by the industry fail to provide sufficient information for this purpose. Approach: This paper explores criteria for effectively planning ALT. 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. Results: We find that the importance of selecting an experimental design exceeds the importance usually given to the selection of sample size and censoring scheme, or to the choice of a test unit allocation strategy in the context of a reliability test case study. D-optimal-based designs provide good performance for acceleration model validation, especially when the testing feasibility region is asymmetric. D-optimal designs offer an eightfold reduction in sample size over legacy designs for a fixed precision. Alternatively, D-optimal designs yield a 3.5x increase in parameter estimate precision over legacy designs for a fixed sample size. Numerical simulations confirm that legacy designs provide insufficient precision to determine if two or three model parameters are significant.