Optimizing the Cleaning Strategies to Reduce the Flood Risk Increased by Gully Blockages

The blockages in gully systems increase the flood risk in urban areas because less surface water can be drained into the sewer pipes beneath. Since a gully system comprises three components of street inlets, gully channels, and lateral tubes, it has been challenging to study the cleaning strategies when multiple gully components are simultaneously blocked, especially when considering the complex interactions in hydraulics and optimization. In this study, we adopted a sophisticated hydraulic model to simulate water transportation across surface, gully, and sewer layers under 16 scenarios with different blockage ratios of gully components for a rainstorm event in 2017 in Taipei City, Taiwan. The results are regressed and served as objective functions in nonlinear programming to obtain optimal cleaning strategies for each gully component that minimize surface floodwater volumes under different frequencies and budgets. The regression results show that floodwater volumes increase the most with the blockage of gully channels compared to the blockages of other components; however, the optimization analysis indicates that street inlets should be cleaned first for having the lowest cost. Compared with the increase of budget, the increase of cleaning frequency is more cost-effective because it prevents the nonlinear increase of floodwater volume at high gully blockage ratios. This research provides a guideline for the cleaning of a gully system in which the blockages of different components must be removed in a sequential manner in order to minimize flood risk.