A BAYESIAN-ANALYSIS OF THE LOGARITHMIC-POISSON EXECUTION TIME MODEL-BASED ON EXPERT OPINION AND FAILURE DATA

In this paper, we propose a Bayesian approach for predicting the numbe r of failures in a piece of software, using the logarithmic-Poisson mo del, a nonhomogeneous Poisson process (NHPP) commonly used for describ ing software failures. A similar approach can be applied to other form s of the NHPP. The key feature of our approach is that now we are able to use, in a formal manner, expert knowledge on software testing, as for example, published information on the empirical experiences of oth er researchers. This is accomplished by treating such information as e xpert opinion in the construction of a likelihood function which leads us to a joint distribution. Our procedure is computationally intensiv e, but for the case of the logarithmic-Poisson model has been codified for use on a personal computer. We illustrate the working of our appr oach via some real live data on software testing. We wish to emphasize that our aim is not to propose another model for software reliability assessment. Rather what we have here is a methodology that can be inv oked with existing software reliability models.