A finite queuing model with generalized modified Weibull testing effort for software reliability

This study incorporates a generalized modified Weibull (GMW) testing effort function (TEF) into failure detection process (FDP) and fault correction process (FCP). Although some researchers have been devoted to model these two processes, the influence of the amount of resources on lag of these two processes is not discussed. The amount of resources consumed can be depicted as testing effort function (TEF), and can largely influence failure detection speed and the time to correct a detected failure. Thus, in this paper, we will integrate a TEF into FDP and FCP. Further, we show that a GMW TEF can be expressed as a TEF curve, and present a finite server queuing (FSQ) model which permits a joint study of FDP and FCP two processes. An actual software failure data set is analyzed to illustrate the effectiveness of proposed model. Experimental results show that the proposed model has a fairly accurate predication capability.