Risk-based optimization of large flood-diversion systems using genetic algorithms

Abstract

This article presents a robust and efficient risk-based genetic-algorithm model for the optimal design of large, temporary flood-diversion systems, in which the routing effect may not be disregarded. This article integrates the flood-routing process and uncertainties in flood-magnitude estimator, as well as the hydraulic uncertainties, into an optimization model. A modification to parameter uncertainty modelling is proposed that is verified using Monte Carlo simulation technique. System design capacity and dimensions are explicitly treated as decision variables. The performance of the model is demonstrated using a hypothetical case example. Furthermore, a series of sensitivity analyses are conducted to assess the effect of uncertainties in damage cost and construction time on the final results. The results indicate that these factors, as well as consideration of flood routing, could have a significant effect on the optimum design capacity of the system.