Software Reliability Growth Modeling Involving Burr Type XII distribution and Fault Removal Efficiency

Abstract

In software reliability analysis various authors have used Burr type XII distribution to model the failure pattern of the system due to its wide variety of flexible shapes. In particular cases it can be reduced to Exponential, Normal, Weibull, Log-logistic, Gamma distributions etc. In proposed paper software reliability growth model has been developed incorporating fault removal efficiency (FRE) and Burr type XII based testing effort function. FRE represents fraction of detected faults which are removed completely. Parameters of model are predicted by LSE whereas MSE is used to perform comparison analysis. Results validate better fitting of data set.