Enhancing the effectiveness of urban drainage system design with an improved ACO-based method

In the context of climate change and urbanization, urban floods have been one of the major issues around the world, causing significant impacts on the society and environment. To effectively handle these floods, an appropriate design of the urban drainage system (UDS) is highly important as its function can significantly influence the flooding severity and distribution. In recent years, evolutionary algorithms (EAs) have been increasingly used to design UDS due to their great ability in identifying optimal solutions. However, low computational efficiency and low solution practicality (i.e. the final solutions do not satisfy the design criteria) are major challenges for the majority of EA-based methods. To this end, this paper proposes an improved ant colony optimization (ACO, a typical type of EAs) based method to enhance the UDS design effectiveness, where the optimization efficiency is enhanced by initializing the ACO using an approximate design solution identified by the engineering design method, and the solution practicality is improved by explicitly accounting for the design criteria within the optimization using a proposed sampling method. The utility of the proposed method is demonstrated using two real-world UDSs with different system complexities. Results show that the proposed method can identify design solutions with significantly improved efficiency and solution practicality compared to the traditional design approach, with advantages being more prominent for larger UDS design problems. The proposed method can be used by researchers/ practitioners to explore and develop better understanding of the UDS design alternatives under various challenges of climate change and rapid urbanization.