Class Thresholds Pre-Definition by Clustering Techniques for Applications of ELECTRE TRI Method

The sorting problem in the Multi-criteria Decision Analysis (MCDA) has been used to address issues whose solutions involve allocation of alternatives classes. Traditional multi-criteria methods are commonly for this task, such as ELECTRE TRI, AHP-Sort, UTADIS, PROMETHEE, GAYA, etc. While using these approaches perform procedure, decision-makers define profiles (thresholds) classes compare within profiles. However, most applications based on subjective tasks, i.e., decision-makers’ expertise, which sometimes might be imprecise. To fill that gap, paper, a comparative analysis method TRI and clustering algorithms is performed obtain an auxiliary procedure initial thresholds method. In proposed methodology, K-Means, K-Medoids, Fuzzy C-Means algorithms, Bio-Inspired metaheuristics PSO, Differential Evolution, Genetic algorithm tested considering dataset from fundamental Water Distribution Networks. computational performances indicate was more suitable achieving desired response. practical contributions show relevant provide view boundaries datasets specific applications. Theoretically, it new development pre-define limits approach.