BAYES PREDICTIVE ANALYSIS OF A FUNDAMENTAL SOFTWARE-RELIABILITY MODEL

The concepts of Bayes prediction analysis are used to obtain predictive distributions of the next time to failure of a software when its past failure behaviour is known. The technique is applied to the Jelinski-Moranda softwarereliability model which in turn can show an improved predictive performance for some data sets even when compared with some more sophisticated software-reliability models. A Bayes software reliability model has been presented which can be applied to obtain the next time to failure pdf and Cdf for all testing protocols. The number of initial faults and the per-fault failure rate are assumed to be s-independent and Poisson and gamma distributed respectively. For certain data sets, the present technique yields better predictions than some alternative methods if the prequential likelihood and U-plot criteria are adopted. As with all models in software reliability, the present one is a further tool which should be tried for any given data set in the quest for a suitable description of software failure-behaviour. A FORTRAN 77 program based on this method which uses the NAG subroutine library is available from the author. © 1990 IEEE