Abstract
In life test experiments it is a common practice to terminate experiment when either a certain number of failures occurs or certain stipulated time is elapsed. Boardman and Kendell (1970) have considered the problem of estimation with respect to type I censoring when an item is subjected to only one of the two causes of failure assuming exponential model. In this paper an extension of the compound exponential model with different failure rates at each stage of censoring is developed when an item is subjected to only one of the two causes of failure. The maximum likelihood estimates alongwith their standard errors are obtained