Pedestrian Performance Analysis and Optimization of Floor Plan Types in Underground Shopping Malls Using Agent-Based Simulation

Underground shopping malls in Korea, despite benefiting from steady pedestrian inflows from urban rail systems, are experiencing rising vacancy rates and declining profitability due to the growth of online retail and decentralized commerce. Conventional revitalization measures—such as marketing events and rent subsidies—have yielded only temporary effects, indicating that structural design may play a more decisive role in long-term viability. This study examines the relationship between floor plan configuration and pedestrian performance, taking Bujeon Mall in Busan as a case study. An agent-based simulation was conducted, modeling four pedestrian types—Passing, Pragmatic, Meandering, and Appropriation—under both single-type and mixed-type conditions, across average and peak density environments. Results demonstrate that spatial configuration exerts a significant influence on movement efficiency, congestion resilience, and spatial utilization. The conventional three-row layout consistently underperformed, especially under peak loads, whereas central openness and zigzag arrangements improved adaptability and promoted more balanced density distribution. Mixed-type scenarios revealed synergistic effects, frequently outperforming single-type conditions. These findings suggest that underground mall regeneration depends not only on physical accessibility but also on cognitive clarity and spatial complexity. Finally, functional strategies are proposed for transit-oriented, commercial, and mixed-use contexts, offering design directions for the sustainable renewal of underground malls.