LAUSR

Abstract Design/rehabilitation of urban stormwater drainage systems has become a challenging issue due to increasing frequency and severity of floods in urbanized areas. Optimization frameworks can provide a proficient computational tool for stormwater management. In this study, using four different optimization algorithms and EPA-SWMM ( E nvironmental P rotection A gency - S torm W ater M anagement M odel) software, a coupled numerical and optimization model was developed to rehabilitate the drainage system in eastern Tehran, Iran. The current drainage network suffers from a significant lack of hydraulic capacity. Thus, combinations of relief tunnels and/or storage units were evaluated and optimal rehabilitation strategies were suggested according to minimizing conflicting objective functions of costs and flooding. Results have revealed that AMALGAM ( A M ulti- AL gorithm, G enetically A daptive M ulti-objective) outperformed three other algorithms, NSGA-II ( N on-dominated S orting G enetic A lgorithm-II), NSHS ( N on-dominated S orting H armony S earch), and NSDE ( N on-dominated S orting D ifferential E volution) for the evaluation of rehabilitation of Urban Stormwater Drainage Systems (USDSs) in terms of convergence and diversity criteria.