A Decision-Support Framework Integrating ISM, Modified Fuzzy MICMAC, and Optimization for Software Requirement Dependency Analysis

Abstract

Several studies have developed methods to model and analyze dependencies among requirements in software development projects. However, existing approaches provide limited support for decision-making related to software requirement (SR) changes. Extending this line of research, the present study proposes a three-stage decision-support framework to more systematically model, analyze, and manage dependencies among software requirements (SRs). First, Interpretive Structural Modeling (ISM) is employed to organize SRs into a simple hierarchical form. Next, a modified version of the Fuzzy Cross-Impact Matrix Multiplication Applied to Classification (MICMAC) analysis, which considers both the likelihood of occurrence and the impact of change, is employed to categorize SRs based on their criticality. Building on the outputs of the ISM and fuzzy MICMAC analyses, a lexicographic multi-objective optimization model is then developed to support the selection of SR changes that satisfy customer requirements while minimizing the breadth and intensity of change propagation. The applicability of the proposed framework is demonstrated through a real-world software development project involving 28 SRs. This demonstration illustrates that the proposed framework provides a systematic and robust decision-support tool for software requirement change management, helping reducing the risk of project cost and schedule overruns.