A hybrid MCDM model with Monte Carlo simulation to improve decision-making stability and reliability

Abstract

Employing an appropriate method to achieve a reliable decision remains a challenge for decision-makers (DMs) in the multiple-criteria decision-making (MCDM) process owing to its inherent model-related complexity, sensitivity, and uncertainty. In this context, this study proposes an innovative hybrid MCDM model that integrates criteria importance through intercriteria correlation (CRITIC), multi-attributive border approximation area comparison (MABAC), and k-means with Monte Carlo simulation (i.e., CRITIC–MABAC–Kmeans with Monte Carlo simulation), aiming to address MCDM problems with substantial stability and reliability. Specifically, MABAC attests to the stability of this method, as it is less affected by normalization and weighting schemes. In addition, the challenge of conflicting k-means clustering outcomes, owing to diverse initial centroid selections, is mitigated by a Monte Carlo simulation, which identifies the most probable type of result and compensates for small-sample size bias. The model performance is tested using a case study of observing transport safety accomplishments in the ASEAN region. Enhanced multiple comparisons of the experimental results verify the quality, efficiency, and adaptability of the proposed model, indicating its feasibility for DMs, policymakers, and practitioners as a practical tool for handling real-life MCDM activities in various domains under compounded sensitivity and uncertainty.