Evaluating contractor performance in water projects using k-means, hierarchical methods, and fuzzy AHP

• The study uses FAHP and clustering methods to evaluate 37 water contractors, improving decision-making. • The study highlights the need for multi-criteria methods to assess contractors on performance criteria like response times. • PCA was used to visualize clustering results, clarifying contractor performance distributions and identifying outliers. • Fuzzy AHP ranked contractor R1 as the top performer, followed by R14 and R16, showing the value of expert judgment. • Clustering identified six outliers (R14, R1, R16, R19, R25, R27), indicating significant performance deviations needing investigation. Contractor performance evaluation in water-infrastructure projects is a multi-criteria and uncertainty-prone task. This study applies two analytical paths to assess 37 water-distribution contractors using six ratio-based performance criteria. First, Fuzzy Analytic Hierarchy Process (FAHP) is used to weight the criteria under pooled expert judgments and to produce an uncertainty-aware contractor ranking. Second, K-means and Agglomerative clustering are applied independently to the same weighted, standardized performance data to identify similarity-based contractor groups and stable outliers. The clustering results consistently revealed six outliers (R14, R1, R16, R19, R25, and R27), indicating contractors with performance profiles that deviate sharply from the majority. FAHP ranked R1 as the top contractor, followed by R14 and R16, while R27 placed 34th. Comparing the independent clustering structure with the FAHP ordering provides a triangulated interpretation that validates high- and low-performance patterns and highlights exceptional or problematic contractors for managerial attention. The proposed FAHP–clustering framework strengthens robustness and transparency in contractor evaluation and supports more reliable decision-making in project management.