.png){kind=icon}
.png){kind=icon}
Return
An Improved TOPSIS Model Based on Weighted Generalized Mahalanobis Distance for Multi-Attribute Decision-Making in Water Reservoir Operations
.png){kind=icon}
PRE
AI
S
P
DOI：10.1007/s11269-026-04833-7.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Managing large-scale hydraulic projects involves complex trade-offs between flood control, power supply, and ecological sustainability. Although the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is widely used for multi-attribute decision-making, the traditional Euclidean distance formulation fails to account for correlation and multi-collinearity among water resource index attributes. To resolve this limitation, this paper introduces an improved TOPSIS model that replaces Euclidean distance with a weighted generalized Mahalanobis distance metric. This mathematical enhancement captures the underlying coupling relationships between inter-dependent criteria, eliminating parameter redundancy. The proposed model was implemented to evaluate and optimize operational scheduling schemes for a real reservoir project. The multi-attribute evaluation matrix was systematically solved, and the resulting alternative rankings were validated against field datasets. The comparative results demonstrate that the improved model successfully mitigates the ranking inversion flaws inherent to classical distance metrics, delivering a more stable and objective decision-making framework. This unified algorithmic approach provides reservoir managers and engineering consultants with a high-fidelity, intelligent optimization tool to maximize project value, ensure structural safety, and balance economic-ecological constraints in active hydrological systems. An equilibrium model balances period, cost, and quality for multi-mode reservoir construction. IBBPSO algorithm successfully converges on 14 non-dominated Pareto-optimal project schemes. MATLAB criteria correlation matrix identifies heavy dependencies between duration, cost, and quality. Weighted generalized Mahalanobis distance replaces Euclidean TOPSIS to eliminate collinearity. Sensitivity analysis of preference index λ validates ranking stability and model rationality.
Keywords:
Reservoir Construction Management
Multi-attribute decision-making (MADM)
Improved TOPSIS
Generalized Mahalanobis distance
Reservoir operational scheduling
Water resource optimization
Criteria multi-collinearity
Journal
IF:
4.7
Papers:
8.1K
Citations:
1.6W
Organization
Cited Papers