An Improved Software Reliability Model Considering Fault Correction Rate

Software reliability growth models (SRGMs) base on NHPP are quite successful. Various NHPP-SRGMs have been built upon various different assumptions, for example, software failure rate, the total number of faults, and the fault detection rate and so on. But in fact, the fault correction rate is not a constan, but a time function. In this paper, we aim to incorporate fault detection rate, the number function of faults and fault correction rate into conditional SRGMs. Experimental results indicate that the new model which proposed in this paper has a fairly accurate prediction capability.