Abstract
The increasing deployment of algorithmic decision-making systems in urban governance promises efficiency but raises concerns about equity, particularly in resource allocation. This study investigates how algorithmic resource allocation affects equity across socioeconomic and demographic groups in a mid-sized city. We developed a simulation model integrating GIS data, demographic indicators, and algorithmic allocation rules to test equity outcomes under different transparency and fairness constraints. Results show that unconstrained optimization algorithms favor high-demand, high-income areas, widening existing disparities. Incorporating equity-aware constraints—such as minimum allocation floors for disadvantaged neighborhoods—reduces inequality by 34% without significant efficiency losses. However, transparency mechanisms alone do not mitigate bias without explicit fairness objectives. We also find that human oversight can introduce inconsistencies, suggesting a need for hybrid decision-making frameworks. The findings contribute to smart urban governance by providing empirical evidence on algorithmic equity and offering policy recommendations for equitable resource allocation.