Systematic review of software engineering uses of multi-criteria decision-making methods: Trends, bibliographic analysis, challenges, recommendations, and future directions

Correctly adhering to the processes within the software development life cycle (SDLC), from analysis and design to coding and testing, is vital for ensuring the successful and efficient creation of high-quality software applications. These structured phases provide a systematic approach to software development, facilitating clear communication, reducing errors, and improving collaboration among development teams. For the proper and correct use of SDLC processes, it is essential for both software engineers and software decision makers to perform the correct and needed actions while performing each software process, and one method of facilitating that is multicriteria decision making (MCDM). This study aims to provide a systematic review of the use of MCDM within the field of software engineering (SE), encompassing methodologies such as fuzzy MCDM, AHP, TOPSIS, DEMATEL, and other methods, with a deliberate focus on software engineering development processes. To ensure the high quality of this review, a methodical and structured literature search process was performed with strict selection criteria, resulting in the identification of 32 contributions on the applications of MCDM in SE from various databases, including Scopus, ScienceDirect, IEEE Xplore digital library (IEEE), and Web of Science (WOS). The selected papers were taxonomized into seven main categories, with some divided into subcategories. This paper presents a systematic and comprehensive analysis of the aforementioned studies, investigating the challenges, motivations, and recommendations found within each, thereby paving the way for potential future research. Bibliometric analysis is also provided to show concise quantitative analysis of related bibliographic information, which draws several key insights into publication trends. Finally, a critical analysis of the current literature and existing research is presented, while also addressing relevant research gaps.