A Novel Risk Programming Model for Water Resources Management under Uncertainty

Risk aversion is one of the most important issues in water resources management which may become more complex with existence of uncertainty. To deal with this kind of problems, a novel risk programming model, CNTPS_CRI, was proposed in this paper. It was formulated through integrating two models: CVaR-based nonlinear stochastic programming model (CNTSP), and comprehensive risk index (CRI). The CNTSP is formulated through integrating a Conditional Value-at-Risk (CVaR) model into a two-stage stochastic programming (TSP) framework. It could be used to analyze pro-defined policy scenarios and deal with random uncertainties. Moreover, measurement of extreme expected loss on the second-stage penalty cost was incorporated into the model constraints, such that the trade-off between system economy and extreme expected loss could be accounted for. And the nonlinear penalty function avoid the situation that single water users to undertake most of expected risk loss. To help decision maker to choose a suitable risk-aversion alternatives among obtained solutions, the CRI is employed consequently. The developed CNTPS_CRI was applied to a water resources allocation problem. The results demonstrated that the proposed model could help decision makers obtain effective and reasonable allocation alternatives, and make a reasonable decision.