David D. Hanagal - Software Reliability Growth Models

The Infosys Science Foundation Series in Mathematical Sciences, a Scopus-indexed book series, is a sub-series of the Infosys Science Foundation Series. This sub-series focuses on high-quality content in the domain of mathematical sciences and various disciplines of mathematics, statistics, bio-mathematics, financial mathematics, applied mathematics, operations research, applied statistics and computer science. All content published in the sub-series are written, edited, or vetted by the laureates or jury members of the Infosys Prize. With this series, Springer and the Infosys Science Foundation hope to provide readers with monographs, handbooks, professional books and textbooks of the highest academic quality on current topics in relevant disciplines. Literature in this sub-series will appeal to a wide audience of researchers, students, educators, and professionals across mathematics, applied mathematics, statistics and computer science disciplines.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.

The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Software plays an important role in both real life and industrial organizations. In order to assure the reliability of software, a long testing process is usually needed before the software can be finally released to market. During the last few years, several software reliability growth models (SRGMs) have been introduced to characterize the growth of software reliability during testing phase. These models are very important for management of the orgaanizations to make accurate decisions, such as determining the optimal software release time considering both software reliability and total cost.

These models were developed based on the assumption that faults detected in the testing phase are removed immediately with no debugging time delay and no new faults are introduced into the software. In other words, it is assumed that whenever an attempt is made to remove a fault, it is removed with certainty and this is referred as perfect debugging. But the debugging activity is not always perfect because of a number of factors like the testers skill and expertise. The testing team however, may not be able to remove a fault with certainty when a software fault is observed and the original fault may remain, leading to a phenomenon known as imperfect debugging. Another possibility is that while correcting a software error additional errors may be generated and these errors may get into the software. Such models may be referred as error generation models. In the case of error generation, the total fault content increases as testing progresses because new faults are introduced into the system while removing the original faults.

In this book, we have discussed different software reliability models which include both increasing/decreasing nature of hazard function. Also we have discussed software reliability growth models incorporating the notion of error generation over time as an extension of delayed S-shaped software reliability growth model. Further actual failure data sets are applied to these models and a comparison study is carried out using goodness of fit measures. Research activities in software reliability engineering have been conducted and a number of NHPP software reliability growth models have been proposed to assess the reliability of software. In fact, software reliability models based on the NHPP have been quite successful tools in practical software reliability engineering. These models consider the debugging process as a counting process characterized by its mean value function. We obtain software reliability function and the mean value function. Model parameters are usually estimated using either the maximum likelihood method or regression. Different models have been built upon different assumptions.