LOUD has published research titled "Classification of Tokyo Urban Parks Based on Mobile Big Data of Visitor Behavior" in the journal Applied Geography

Shanghai Key Laboratory of Urban Design and Urban Science (LOUD) published a paper titled "Visitation-Based Classification of Urban Parks Through Mobile Phone Big Data in Tokyo" in the journal Applied Geography.

Figure: Study Area and Park Locations

Figure: Clustering Results and Spatial Distribution of Parks Based on Visitation Behavior

Urban parks play a crucial role in promoting physical activity, mental health, and environmental conservation, forming a core component of green infrastructure planning. While geographic coordinate reference data have advanced the study of park green spaces, existing park classifications often overlook actual visitation behavior patterns. This study reclassifies urban parks in Tokyo based on over 5.9 million records from approximately 330,000 visitors to 300 parks, comparing the results with size-based park classifications.We employed a series of analytical methods, including principal component analysis (PCA), Isolation Forest, clustering algorithms, and the Gini coefficient. The results reveal four key visitation metrics: activity intensity, usage efficiency, time occupancy, and revisit rate. These metrics identified parks with atypical visitor patterns within each size category, leading to three new park classifications: everyday leisure parks, social destination parks, and seasonal activity parks.Additionally, we found significant disparities in travel distances to parks, particularly at night, on weekends, and holidays, with pronounced inequalities in seasonal activity parks and small parks. The findings suggest tailored park management strategies, prioritizing maintenance and facility development based on observed visitor patterns to enhance recreational potential. This study provides insights for urban park research, supporting the development of green infrastructure planning and policies aimed at improving park usage and enjoyment.

Dr. Zhou Yichun, one of the authors of the paper, pointed out that urban parks play a crucial role in promoting residents' physical and mental health. Existing park classification systems often overlook actual visitation patterns, making them insufficient to fully reflect park usage and potential demand. By classifying parks based on visitor behavior patterns, the study not only reveals the actual usage of different types of parks but also helps urban planners and park managers allocate resources more effectively, optimize park facilities, and enhance the recreational and social functions of parks.In his future research, Dr. Zhou plans to integrate social perception and artificial intelligence technologies to expand the study area and data sources. This will provide a more comprehensive understanding of how parks are used and support the development of more scientific and effective management strategies.

Professor Ying Li emphasized that this research not only provides a new theoretical basis for the planning and management of urban parks and green spaces but also reveals the diversity and complexity of park usage patterns through empirical analysis. The study is an important component of the laboratory project titled "Urban Perception AI Technology-Supported Blue-Green Infrastructure Planning and Design" and represents a significant exploration of digital practices in urban parks.Through this research, it is hoped to attract more scholars and experts, both domestically and internationally, to collaboratively participate in the study and design of future urban blue-green infrastructure. This initiative aims to provide new perspectives and methods for addressing climate change and enhancing urban livability.

Professor Wu Longfeng from Peking University pointed out that understanding how residents use urban parks and green spaces is one of the key ways to promote their health benefits and improve urban environmental quality. This study uses cities in developed countries as case studies and employs diverse big data to extract universal patterns of how residents use parks, closely integrating these findings with park green space planning, design, and urban management policies, which holds significant theoretical and practical implications.Professor Wu stated, "This research is one of the collaborative outcomes between the Planning Department of the School of Urban and Environmental Sciences at Peking University and LOUD. Both sides complement each other’s strengths, combining our research perspectives on 'urban blue-green space planning and public health' with the laboratory's expertise in 'urban perception and artificial intelligence' technologies. This collaboration lays a solid foundation for further joint research."

Professor Jihoon Song from Myongji University in South Korea stated that after more than five years of long-term collaboration, researchers from different countries and institutions have discovered that cooperation is key to creativity and productivity. Scholars from three different Asian countries have just published their third paper in a core journal and plan to publish more in the future. Regarding the utilization of urban parks, this study introduces a new analytical framework based on "big data" and reveals significant differences between reality and traditional planning practices. "I hope this collaborative research will ultimately validate the scientific basis of urban planning in Asia while providing valuable lessons for other regions," Professor Song remarked.

Professor Yuki Akiyama from Tokyo City University pointed out, " LOUD and our lab continue to collaborate on research utilizing mobile phone big data. Our lab mainly focuses on the fundamental processing of pedestrian flow big data in Tokyo and provides the technical expertise to clean the data to meet the needs of academic collaboration. Through our partnership with LOUD, we aim to explore new possibilities in pedestrian flow big data and provide useful information for urban and transportation planning. Since both laboratories have similar research areas and social practice goals, we hope to collaborate on various other research projects."

Paper Link:

Zhou, Y., Guan, C., Wu, L., Li, Y., Nie, X., Song, J., Kim, S. K., & Akiyama, Y. (2024). Visitation-based classification of urban parks through mobile phone big data in Tokyo. Applied Geography, 167, 103300. https://doi.org/10.1016/j.apgeog.2024.103300