Akademik Veri Yönetim Sistemi
IEEE Access, 2026 (SCI-Expanded, Scopus)
Dockless micromobility systems provide essential flexibility for urban mobility but face significant operational challenges, primarily the supply-demand imbalance where vehicles are unavailable in high-demand areas while remaining idle elsewhere. Although significant research has been conducted in the literature to address this imbalance using spatial optimization models such as the Maximal Covering Location Problem (MCLP), they often fail to accurately model users' walking behavior to access a vehicle. Specifically, existing approaches neglect the combined influence of distance and topographical factors such as slope, leading to suboptimal vehicle placement. This paper proposes a novel Walking Behavior-Based MCLP (WB-MCLP) model. This approach integrates a more realistic representation of user behavior by simultaneously modeling walking characteristics as a function of both distance and slope to optimize vehicle deployment. The proposed WB-MCLP was validated using historical ride data from stationless shared e-scooter systems and benchmarked against existing models, including Distance Tolerance MCLP (DT-MCLP) and slope-aware variant (DT-MCLP+S), as well as genetic algorithm (GA) and random placement (RP) strategies. Results demonstrate that the WB-MCLP achieves superior performance in 14 out of 20 tested scenarios in terms of walking distance and 11 out of 20 scenarios in terms of coverage rate, providing more effective vehicle coverage that aligns with realistic user access behavior. The findings confirm that incorporating both distance and slope provides a more accurate model of user behavior, leading to significant improvements in system efficiency. This research offers a valuable decision support framework for service providers and urban planners to enhance the performance and sustainability of dockless shared mobility systems.