An optimization model for water resources allocation risk analysis under uncertainty
In order to deal with the risk of low system stability and unbalanced allocation during water resources management under uncertainties, a risk-averse inexact two-stage stochastic programming model is developed for supporting regional water resources management. Methods of interval-parameter programming and conditional value-at-risk model are introduced into a two-stage stochastic programming framework, thus the developed model can tackle uncertainties described in terms of interval values and probability distributions. In addition, the risk-aversion method was incorporated into the objective function of the water allocation model to reflect the preference of decision makers, such that the trade-off between system economy and extreme expected loss under different water inflows could be analyzed. The proposed model was applied to handle a water resources allocation problem. Several scenarios corresponding to different river inflows and risk levels were examined. The results demonstrated that the model could effectively communicate the interval-format and random uncertainties, and risk aversion into optimization process, and generate inexact solutions that contain a spectrum of water resources allocation options. They could be helpful for seeking cost-effective management strategies under uncertainties. Moreover, it could reflect the decision maker's attitude toward risk aversion, and generate potential options for decision analysis in different system-reliability levels.