Flood?Resilient Urban Road Networks: Integrating Hydrologic Forecasting, Dynamic Traffic Diversion, and Socio Economic Equity into a Multi Criteria Optimization Framework under Deep Climate Uncertainty

Abstract

Intensifying rainfall extremes routinely overwhelm urban drainage systems, submerging road corridors and paralyzing essential services. This study proposes a holistic decision framework that couples scenario based hydrologic projections with network level optimization to raise the flood resilience of city streets. A multi criteria genetic algorithm simultaneously minimizes expected travel delay, emergency?response accessibility loss, and retrofit cost while maximizing equity of service to vulnerable neighborhoods. Comparative simulations on a 482-km road grid from a monsoon prone Indian metropolis reveal that targeted kerb height adjustments, porous pavement retrofits, and deployable flood-gates on only 7?% of links can reduce two hour post storm congestion by 54?% and shrink ambulance detours by 38?%. Sensitivity analysis across six Representative Concentration Pathway ensembles demonstrates that benefit cost ratios remain above 2.3 even under a 30?% precipitation escalation. The framework offers practitioners a transparent tool to prioritise incremental, fiscally feasible upgrades while acknowledging the deep climatic and socio-economic uncertainties that shape urban mobility futures.