Methodology for integrated multicriteria decision-making with uncertainty: Extending the compromise ranking method for uncertain evaluation of alternatives

Making a decision usually means selecting one from different alternatives to solve a problem according to a set of criteria. Multicriteria analysis usually offers a quantitative approach to ease decision-making by ranking the alternatives. However, uncertainty can arise when rating the importance of criteria and the adequacy of each alternative for each criterion, due to two factors: first, answers are usually expressed in linguistic terms that do not have a unique quantification; and second, there might be a lack of confidence in the response. Most multicriteria procedures combine fuzzy numbers and linguistic scales to deal with the first factor, but underestimate confidence issues. In this context, this work develops a Methodology for Integrated Multicriteria Decision-making with Uncertainty (MIMDU), which considers both factors of uncertainty. MIMDU is structured in three phases: (1) a novel procedure based on fuzzy rating scales to model uncertain opinions; (2) a fuzzy formulation of the compromised ranking method to rank the alternatives; and (3) a systematic procedure for results' interpretation comparing a crisp ranking (without uncertainty) and a fuzzy-based ranking (with uncertainty). The methodology is illustrated with a generic example case, aiming to prove its potential application in any sector. Results show that MIMDU helps decision-makers to choose the most reliable alternative, since significant differences in ranking with and without uncertainty can be addressed. A sensitivity analysis is carried out to bare the effect of confidence in the alternatives evaluation, concluding that worse rankings are obtained for alternatives that are less confidently evaluated. A final comparison with the standard fuzzy VIKOR method shows MIMDU's major preciseness in modelling non-confident opinions and providing more useful and complimentary information to better assist decision-making.(c) 2021 Elsevier B.V. All rights reserved.