| Article ID: | iaor20123884 |
| Volume: | 14 |
| Issue: | 2 |
| Start Page Number: | 200 |
| End Page Number: | 220 |
| Publication Date: | May 2012 |
| Journal: | International Journal of Operational Research |
| Authors: | Mellouli Khaled, Argoubi Majdi, El Kamel Abdelkader |
| Keywords: | decision, programming: dynamic, stochastic processes, control processes |
A decision support system (DSS) has been developed for integrated, real‐time flood control in a multi‐reservoir system. The DSS integrates an optimisation module with a geographic information system, and a graphical dialogue interface allowing effective use by system operators. A neuro‐dynamic programming technique, which overcomes numerous limitations of stochastic dynamic programming, is introduced for determining the optimal release policy under flood conditions. The method is illustrated by a real‐world case study.