A New Homogeneous Pure Birth Process Based Software Reliability Model

A new Software Reliability model based on a pure birth process is proposed. In our novel approach, the birth (failure) rate of the process is considered to be non dependent on time but dependent non linearly on the previous number of births (failures), contrarily to non homogeneous pure birth processes, as it is usually done in the literature. We use the empirical Bayes framework in order to get the birth (failure) rate. Our approach allows either to estimate the mean time to failure MTTF or to simulate the stochastic failure process. We analyze both, the discrete and continuous time case and apply the first to a real case.