Sensitivity analysis approaches in multi-criteria decision analysis: A systematic review

• Published in: Applied soft computing Vol. 148; p. 110915
• Main Authors: Więckowski, Jakub, Sałabun, Wojciech
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.11.2023
• Subjects: Decision making; Literature review; MCDA; PRISMA; Sensitivity analysis; Sensitivity analysis frameworks; Sensitivity analysis frameworks; Literature review; PRISMA; Sensitivity analysis; Decision making; MCDA
• ISSN: 1568-4946; 1872-9681
• DOI: 10.1016/j.asoc.2023.110915

In the field of Multi-Criteria Decision Analysis (MCDA), the deployment of sensitivity analysis has become a fundamental approach to testing the robustness and reliability of the results obtained. Its importance lies in its ability to gain additional insight into potential changes that affect the desirability of the decision variants being evaluated. Through different methodologies, it reveals the vulnerability of results to changes in the underlying data, thus offering valuable information to support decision making. This research paper highlights a compendium of 250 researches involving the application of sensitivity analysis in a multi-criteria context. Using a structured approach of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology, this study provides a comprehensive review process. The main objectives are to identify and categorize common sensitivity analysis techniques explained in the literature and to describe a selected framework that facilitates the execution of these techniques. The research demonstrates the importance of sensitivity analysis to enhance the credibility of MCDA outcomes and contributes to the field of decision support. •Extensive review on comprehensive analysis of sensitivity analysis in MCDA.•Identifies main categories of sensitivity analysis methods in MCDA.•Review on techniques reveal stability of rankings.•Review on techniques reveal changes in alternatives’ preferences.•Frameworks for conducting sensitivity analysis in MCDA.