Generation of operating rules with stochastic dynamic programming and multiple objectives

Conclusion Several models were presented for finding Pareto-optimal solution for a reservoir operating problem. By simply changing the weights of the objectives, as many operating rules as is desired can be generated. Each result yields an operating rule which is a compromise solution between several noncommensurable objective functions. These results should be further analyzed through simulation on a real-time basis and application of multi-objective techniques for final selection of an optimal operating rule.