To enhance the effectiveness of watershed load reduction decision making, the Risk Explicit Interval Linear Programming （REILP） approach was developed in previous studies to address decision risks and system returns...To enhance the effectiveness of watershed load reduction decision making, the Risk Explicit Interval Linear Programming （REILP） approach was developed in previous studies to address decision risks and system returns. However, REILP lacks the capability to analyze the tradeoff between risks in the objective function and constraints. Therefore, a refined REILP model is proposed in this study to further enhance the decision support capability of the REILP approach for optimal watershed load reduction. By introducing a tradeofffactor （α） into the total risk function, the refined REILP can lead to different compromises between risks associated with the objective functions and the constraints. The proposed model was illustrated using a case study that deals with uncertainty- based optimal load reduction decision making for Lake Qionghai Watershed, China. A risk tradeoff curve with different values of a was presented to decision makers as a more flexible platform to support decision formulation. The results of the standard and refined REILP model were compared under 11 aspiration levels. The results demon- strate that, by applying the refined REILP, it is possible to obtain solutions that preserve the same constraint risk as that in the standard REILP but with lower objective risk, which can provide more effective guidance for decision makers.展开更多
基金This paper was supported by the National Natural Science Foundation of China （Grant No. 41222002）, Research Fund for the Doctoral Program of Higher Education of China （20100001120020） and ＂China National Water Pollution Control Program＂ （2013ZX07102-006）. Special thanks to Dr. Daniel Obenour in University of Michigan.
文摘To enhance the effectiveness of watershed load reduction decision making, the Risk Explicit Interval Linear Programming （REILP） approach was developed in previous studies to address decision risks and system returns. However, REILP lacks the capability to analyze the tradeoff between risks in the objective function and constraints. Therefore, a refined REILP model is proposed in this study to further enhance the decision support capability of the REILP approach for optimal watershed load reduction. By introducing a tradeofffactor （α） into the total risk function, the refined REILP can lead to different compromises between risks associated with the objective functions and the constraints. The proposed model was illustrated using a case study that deals with uncertainty- based optimal load reduction decision making for Lake Qionghai Watershed, China. A risk tradeoff curve with different values of a was presented to decision makers as a more flexible platform to support decision formulation. The results of the standard and refined REILP model were compared under 11 aspiration levels. The results demon- strate that, by applying the refined REILP, it is possible to obtain solutions that preserve the same constraint risk as that in the standard REILP but with lower objective risk, which can provide more effective guidance for decision makers.
