RESEARCH ARTICLE ‐ METHODOLOGY Multicriteria decision aid for agile methods evaluation using fuzzy PROMETHEE Omar El Beggar LIM Laboratory—Advanced Smart Systems, Computer Science Department, Faculty of Sciences and Techniques of Mohammedia, University Hassan II Casablanca, Casablanca, Morocco Correspondence Omar El Beggar, Computer Science Department, Faculty of Sciences and Techniques of Mohammedia, Boulevard Mohamed VI, MailBox 146 Mohammedia, PC 20650 Morocco. Email: elbeggar_omar@yahoo.fr Abstract The last decade is marked by the blossom of agile methods and their growing use in many IT projects. They shared common principles and values, such as customer col- laboration, early and continuous delivery of releases, accept requirements changing, and other common principles advocated by the agile manifesto since 2001. Mean- while, there are significant differences among the practices adopted by the agile methods to fulfill agility's principles and values. The differences could be interpreted in terms of how an agile method judges the relative importance of such principles or its orientation to a specific project viewpoint (organizational, technical, or concep- tual) rather than others. This fact leads to a necessity of a deep evaluation of agile methods to choose the best‐fit method suited for a project. Moreover, vague and uncertain information used to evaluate those lightweight methodologies could repre- sent an additional difficulty for decision makers (DM). To drawback this problem, the multiple criteria decision aid (MCDA) method fuzzy PROMETHEE is proposed to assess agile methods with regard to different conflicting criteria and handling besides uncertainty in decision making. Furthermore, a sensitivity analysis is provided to val- idate results and verify their stability. KEYWORDS agile methods, decision making, fuzzy PROMETHEE, linguistic terms, MCDA 1 | INTRODUCTION Many agile methods are firstly used in industrial projects, but in the last decade, they are increasingly adopted for IT projects, and they know in short term a sharp rise of interest and popularity as new software engineering methodologies. Promoting communication between project stake- holders, customer collaboration, fast delivery at a steady pace of software, and acceptance of change are the principal values of agile software development. 1 Whereas the degree of compliance with these values differs from one method to another, as well as each one has its own practices that might have effects on the project progress. In fact, although agile methods present some interesting features regarding some project points of view, they have also some limitations com- pared with other ones. 2 For instance, the eXtreme Programming (XP) is more focused on technical practices rather than management practices such as SCRUM method. 3 XP is suitable for small and medium projects and cannot support large or complex projects like Feature‐Driven Devel- opment (FDD). 4 Besides, contrary to Scrum and XP which define responsibilities and roles in IT project, Kanban method does not. 2 In other words, even as agile methods share common characteristics and principles advocated by the agile Manifesto 5 in 2001, none of these lightweight methods could be adopted for any kind of IT project. Silva et al 6 claimed that according to the context in which the software will be devel- oped, an agile method is more suitable than others. Therefore, one of the main challenges facing IT project manager is to make the decision of selecting the appropriate agile method from several methodologies for a specific project. Furthermore, involving conflicting criteria to make a decision might have significant consequences on project success as it is claimed by Silva et al. 6 Hesari et al and Taromirad and Ramsin 7,8 argued that a deep Received: 2 May 2018 Revised: 24 July 2018 Accepted: 26 July 2018 DOI: 10.1002/smr.2108 J Softw Evol Proc. 2018;30:e2108. https://doi.org/10.1002/smr.2108 © 2018 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/smr 1 of 21