Cities, on every level, including urban transportation, are being compelled to pick ecological solutions in response to global ecological constraints. As a viable alternative to automobiles powered by fossil fuels inside urban areas, sustainable Bike-Sharing Systems (BSS) have quickly gained prominence as an integral part of the global transportation infrastructure. This work undertakes the dataset from the docking BSS in Seoul, South Korea, which contains the details of the bikes rented from different bike docks. Various clustering algorithms such as Birch, Agglomerative, K-Means, and K-Medoids have been used to cluster the docks based on the usage patterns of the bike in those bike docks. Different indices, namely Silhouette, Calinski-Harabasz, and Davies-Bouldin, have been used to measure the performance of the clustering algorithms. An exploratory data analysis presents that there is a significant difference in rental patterns during different times of the day, dates, and months in a year. In addition, an analysis has been carried out to identify demand patterns and detect over-demand clusters. From the experiments, it is found that K-Means gives a better clustering coefficient, which indicates the demands for bikes at different clusters. The findings provide a practical foundation for priority-based bike redistribution and decision support, enabling more efficient resource allocation in urban mobility systems. The proposed framework offers a scalable and data-driven approach for analyzing large-scale bike-sharing datasets and can be extended toward adaptive and real-time decision-making in future intelligent transportation systems.
Subramanian et al. (2026) studied this question.