Step partially accelerated life tests under finite mixture models

Abstract

In this paper, step partially accelerated life tests are considered when the lifetime of an item under use condition follows a finite mixture of distributions. The analysis is performed when each of the components follows a general class of distributions, which includes, among others, the Weibull, compound Weibull (or three-parameter Burr type XII), power function, Gompertz and compound Gompertz distributions. Based on type-I censoring, the maximum likelihood estimates (MLEs) of the mixing proportions, scale parameters and acceleration factor are obtained. Special attention is paid to a mixture of two exponential components. Simulation results are obtained to study the precision of MLEs.

Keywords:

• finite mixtures
• step partially accelerated life tests
• acceleration factor
• type-I censoring
• maximum likelihood estimation
• local Fisher information matrix
• simulation

Acknowledgements

The remarks and comments of the referee are appreciated.