Agent-Based Model for Urban Administration: A Case Study of Bridge Construction and its Traffic Dispersion Effect

From the perspective of urban administration, simulation can be used as an evaluation tool. Specifically, it can provide an intuition to the current urban situation and quantitatively verify the effectiveness of urban policies. This study proposes a traffic simulation model for Sejong city in South Korea. The proposed model is developed as an agent-based model, which describes the movement behaviors of individual agents representing the whole population in the real city. In particular, to evaluate city-level administrative effects, the proposed model incorporates the multiple distributions of city reality by combining various types of observed real data. By aggregating the individual-level movement behaviors, the proposed model generates the demand for the city's transportation system, and the generated traffic demands were statistically validated with the real data. Based on the secured validity, we conducted a case study where the proposed model was used to compare and analyze the effect of traffic dispersion by taking the policy candidates of new bridge construction into account. From the policy experiment results, we discovered policy implications on an effective bridge construction. Furthermore, we found methodological implications of the urban transport model from the microscopic analysis, which is enabled by the virtue of the proposed model structure.