Selection of optimal software reliability growth models using an integrated entropy-Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) approach

A large number of software reliability growth models (SRGMs) have been studied to estimate the reliability of software systems over the past 40 years. Different models have been developed upon different sets of assumptions. A few models were developed in a practical environment by considering testing effort, testing coverage, time delay fault correction, and fault reduction factor. Generally, SRGMs are not dataset independent, and thus the selection of an appropriate SRGM for use in a specific application is a challenging task in the software reliability area. The wrong selection of SRGMs might create a wrong estimation of reliability and consequently a delay in the release of the software. To overcome this problem, we have proposed a unique hybrid entropy weight-based multi-criteria decision-making (MCDM) method and Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) approach for the selection of a suitable SRGM and applied it for optimal selection and ranking of SRGMs. The proposed hybrid approach identifies the need of the relative importance of criteria for a given application without which inter-criterion comparison cannot be accomplished. It requires a set of model selection criteria along with a set of SRGMs and their level of criteria for optimal selection. It successfully displays the result in terms of a merit value which is used to rank the SRGMs. The proposed approach has been validated on two data sets of the real software failure. The results of the study play a vital role for the decision maker to judge the suitability of the SRGM.