This paper discusses the Bayesian approach to estimation and prediction of the\nreliability of software systems during the testing process. A Non-Homogeneous\nPoisson Process (NHPP) arising from the Musa-Okumoto (1984) software reliability\nmodel is proposed for the software failures. The Musa-Okumoto\nNHPP reliability model consists of two componentsâ??the execution time\ncomponent and the calendar time component, and is a popular model in\nsoftware reliability analysis. The predictive analyses of software reliability\nmodel are of great importance for modifying, debugging and determining\nwhen to terminate software development testing process. However, Bayesian\nand Classical predictive analyses on the Musa-Okumoto (1984) NHPP model\nis missing on the literature. This paper addresses four software reliability issues\nin single-sample prediction associated closely with development testing\nprogram. Bayesian approach based on non-informative prior was adopted to\ndevelop explicit solutions to these problems. Examples based on both real and\nsimulated data are presented to illustrate the developed theoretical prediction\nresults.
Loading....