Efficient Multiobjective Storm Sewer Design Using Cellular Automata and Genetic Algorithm Hybrid

Abstract

Optimal sewer design aims to find cost-effective solutions for designing sewer networks, and genetic algorithms (GAs) are one of the state-of-the-art optimization techniques that have been applied to this problem. However, finding good quality solutions by using a GA can be prohibitively time consuming, especially when designing large networks. This paper introduces an efficient and robust hybrid optimization method, which deals with the design task in a multiobjective optimization manner using two consecutive stages. A localized approach based on cellular automata principles is applied at the first stage to obtain a set of preliminary solutions, which are then used to seed a multiobjective genetic algorithm (MOGA) at the second stage. Two large real sewer networks are tested for case studies. Results clearly show that the hybrid approach can surpass the standard MOGA in terms of optimization efficiency and quality of solutions.