When product reliability at normal use conditions is high, you can test the product under conditions that are more severe than normal to reduce the time until failures occur. 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 more information about nonlinear models, see Nonlinear Models in the Nonlinear Designs section.
The ALT Design platform creates and evaluates designs for situations involving one or two acceleration factors. You can optimize your design using D-optimality, quantile estimation, or failure probabilities.
Creating an ALT design requires initial estimates of the acceleration model parameters. Use subject matter knowledge to define starting estimates. If you do not have information to inform your starting estimates, you can specify a multivariate normal prior distribution to describe their uncertainty.
You can also use the Accelerated Life Test (ALT) Design platform to augment a current experiment. You might want to augment a current design to decrease the variance of estimates.