ABSTRACT
As the multi-criteria decision-making (MCDM) problems become increasingly complex, experts have difficulty in quantifying their opinions on each alternative and prefer expressing their preference information linguistically. Among the existing linguistic representation models, the extended comparative linguistic expressions with symbolic translation (ELICIT) model stands out due to its powerful ability to express the hesitancy of experts and obtain results with higher interpretability and accuracy. Given the importance of the aggregation process in the resolution of MCDM problems, the definition of ELICIT aggregation operators has become a research hotspot. Among them is the ELICIT -OWA operator, since the ordered weighted aggregation (OWA) operator is one of the most classical and popular tools to fuse decision information. However, the weights used in the aggregation are computed from the assumption that the criteria are mutually independent, which may be not realistic in some MCDM problems. In this sense, the Choquet integral does allow for reflecting the interrelationship among elements by means of the definition of a fuzzy measure. Therefore, we propose a novel ELICIT-OWA operator that calculates weights via the Choquet integral, taking full account of the relationships between criteria, in order to obtain a realistic aggregation result. On this basis, a new approach to solving MCDM problems is presented. Finally, an illustrative example is applied to demonstrate the practicality and feasibility of the proposed method.
