Continuous software reliability models

For large scale software systems with a huge number of codes, it may be often useful to regard the fault-counting processes observed in testing phase as continuous-state stochastic processes. In this paper we introduce two types of continuous-state software reliability models; diffusion processes with time-dependent drift and non-stationary gamma wear processes, and compare them with the usual non-homogeneous Poisson processes. We expect that our models can provide the better goodness-of-fit performance than the existing software reliability models in many cases and provide convenient assessment tools to estimate the software reliability.