Generation of Failure Data Fitting a Specified Hazard Function

This paper pertains to tools for evaluating statistical tests on failure data. During the research and evaluation phases associated with reliability testing, it frequently becomes necessary to apply the proposed statistical test to a set of failure data with certain known hazards. The use of actual failure data is seldom satisfactory at this point, since the true data does not represent some pure distribution, but represents a mixture of various distributions. This paper presents a set of computational techniques which generate failure data which fit specified hazard functions.