Predictive Analyses of Logarithmic Non – Homogeneous Poisson Process in Software Reliability Using Bayesian Approach with Informative Priors
Nickson Cheruiyot, Luke Akong’o Orawo, Ali Salim Islam · American journal of mathematics and statistics · 2019
Musa – Okumoto (1984) non-homogeneous Poisson Process (NHPP) software reliability model also known as logarithmic NHPP model is one of the widely used reliability model. The model is based on the assumptions that failures are observed during execution time caused by remaining faults in the software; whenever a failure is observed, an instantaneous effort is made to find what caused the failure and the faults are removed prior to future tests and whenever a repair is done it reduces the number of future faults not like other models. The failure intensity function of this model reduces exponentially with time and the expected number of failures has logarithmic function. The predictive analysis of software reliability model is of great importance for modifying, debugging and determining when to terminate software development testing process. This paper presents some results about predictive analyses for the Musa – Okumoto (1984) NHPP model. Four issues in single-sample prediction associated closely with development testing program are addressed. Bayesian approach based on informative prior was adopted to develop explicit solutions to the problems which arise during software development testing process. Developed methodologies were illustrated using real data in form of time between failures.