A New Insight into Software Reliability Growth Modeling

Several software reliability growth models have been presented in the literature in the last three decades. They have been developed for uniform and non-uniform operational profile. Some of them are flexible whereas others are not. Model selection becomes an uphill task. Of late, some authors have tried to develop a unifying approach so as to capture different growth curves, thus easing the model selection process. Some of these approaches use (a) Random lag function (b) Infinite server queuing theory (c) Hazard rate function. The purpose of this paper is to show that all these approaches are equivalent and further show that hazard rate approach is more general and can handle both Imperfect Debugging and Fault generation. This paper thus provides a new insight into the model development and it is shown that how a wide variety of existing software reliability can be unified.