Overview of Accelerated Life Test Designs

When product reliability at normal use conditions is high, you can test the product under conditions that are more severe than normal so that failures occur more quickly. You can use this accelerated failure data to predict product reliability at normal use conditions.

Such a test is called an accelerated life test (ALT).The factors that are set at higher than normal use levels are called acceleration factors. The models for accelerated life tests are typically nonlinear models. For details about nonlinear models, see Nonlinear Models in Nonlinear Designs.

The ALT Design platform creates and evaluates designs for situations involving one or two acceleration factors (and their interaction). You can optimize your design using D-optimality or two types of I-optimality criteria.

Creating a design requires initial estimates of the acceleration model parameters. Since those parameters are usually unknown, you can specify a multivariate normal prior distribution to describe their uncertainty. To decrease the variance of estimates, you can augment existing experiments in the Accelerated Life Test (ALT) Design platform.