Risk Analysis in Design of a Multi-Section Flood Control Levee System: Assessing the Components Coordination and Probability Distribution

Abstract

The purpose of this study is risk analysis in optimum design of a composite multi-section flood control levee system along a river reach considering several uncertainties, and assessing the coordination between the system and its individual components. A stochastic optimization model was developed in which Latin Hypercube sampling and Cholesky decomposition was used for quantifying uncertainties. The results indicated that the multi-section model is more cost-effective than a single-section in such a way that total cost of developing system was reduced by 38%. It was also found that with the increase of the complexity of the model (one and multi-section), and with the separation of uncertainties (into epistemic and aleatory), the coordination of the system decreases 130% and 30% respectively. Finally, it was found that lower and upper tails of a component histogram may be completely differ from a fitted distribution.