Abstract:
|
To collect the information about the lifetime distribution of a product, a standard life testing method at normal operating conditions is not practical when the product has a substantially long lifespan. Accelerated life tests solve this problem by subjecting the test items at higher stress levels for quicker and more failure data. This paper investigates the optimal stress durations for a simple step-stress accelerated life test under Type-I censoring. The determination of the optimal step durations is discussed under several well-known optimality criteria, including C-optimality, D-optimality, and A-optimality. For an exponential population with a single stress variable, the efficiencies of the optimal designs with flexible step durations and uniform step durations are compared through a numerical study and a real case study.
|