A stochastic programming with imprecise probabilities model for planning water resources systems under multiple uncertainties

In this study, a stochastic programming with imprecise probabilities (SP-IP) model is proposed for planning water resources systems. The SP-IP model is capable of addressing multiple uncertainties in the forms of intervals with random boundaries and imprecise probability distributions. The stochastic optimization model can be transformed into a deterministic equivalence in a straightforward manner. A case study of regional water resources allocation is used to demonstrate the applicability of the proposed model. Results indicate that the total net benefits would be decreased with increased probabilities of occurrence, reflecting a potential trade-off between economic benefits and associated risks. The SP-IP model is also capable of providing a variety of decision alternatives under different scenarios of water policies, which is useful for water managers to formulate an appropriate water management policy in an uncertain environment.