Multicriteria Decision Analysis Coupled with Type-2 Fuzzy Bi-Level Programming for Water Resource Optimization: A Case Study of the Jiulong River Basin, China
Xi Zhang, Lei Jin, Haiyan Fu, Yurui Fan, Ruoyu Yin, Li Wang
Abstract
The uneven spatial distribution and irrational allocation of water resources pose significant challenges to economic development and ecological environment in Jiulong River Basin, Fujian, China. In this study, a Type-2 fuzzy bi-level programming (T2FBL) method was developed to optimize the water resource system in the Jiulong River Basin. Future data for the water resources system were predicted using the back-propagation neural network method. The results were analyzed and evaluated using a new multicriteria decision analysis (MCDA) approach. Additionally, the entropy weight and technique for order of preference by similarity to ideal solution (TOPSIS) methods were combined with MCDA to develop the entropy weight TOPSIS method. With the goal of optimizing the water allocation structure in different regions to alleviate water supply pressure, the proposed model uses an improved fuzzy sorting algorithm to address uncertain parameters in the water resources system and considers the conflicting intersections of decision makers at two levels in a bi-level programming model. The results revealed the following: (1) priority was given to adjusting the water distribution structure in Zhangzhou and Longyan in China while developing secondary industries to promote regional economic development; (2) analysis and evaluation of the results of water allocation using the novel MCDA methodology indicated that the optimal scenario resulted in 51.4% increase in tertiary output; and (3) the calculation results of the T2FBL model were analyzed to establish the relationship between water resource allocation, and economic and environmental benefits, essentially serving as a reference for water resource planning. Moreover, this model reduced wastewater discharge by up to approximately 8.2% compared with the fuzzy single-level programming and bi-level programming models.