Acta Geodaetica et Cartographica Sinica ›› 2024, Vol. 53 ›› Issue (8): 1644-1655.doi: 10.11947/j.AGCS.2024.20230258

• Cartography and Geoinformation • Previous Articles     Next Articles

Spatio-temporal cross K-function for geographical flows

Mengjie ZHOU1,2,3(), Mengjie YANG1(), Huiying CHEN1, Yumeng TIAN1, Yiliang WAN1, Jizhe XIA4,5   

  1. 1.School of Geographical Sciences, Hunan Normal University, Changsha 410081, China
    2.Hunan Key Laboratory of Geospatial Big Data Mining and Application, Changsha 410081, China
    3.Key Laboratory for Urban-Rural Transformation Processes and Effects at Hunan Normal University, Changsha 410081, China
    4.School of Architecture and Urban Planning, Shenzhen University, Shenzhen 518060, China
    5.Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ), Shenzhen 518060, China
  • Received:2023-06-08 Published:2024-09-25
  • Contact: Mengjie YANG E-mail:mengjiezhou@hunnu.edu.cn;mengjiezhou@hunnu.edu.cn;yangmengjie@hunnu.edu.cn
  • About author:ZHOU Mengjie (1990—), female, PhD, associate professor, majors in geographical information mining and visualization. E-mail: mengjiezhou@hunnu.edu.cn
  • Supported by:
    The National Natural Science Foundation of China(41901314);The Natural Science Foundation of Hunan Province(2023JJ40447);The Scientific Research Project of Hunan Provincial Department of Education(23B0093);The Open Research Fund Program of Guangdong Key Laboratory of Urban Informatics(GEMlab-2023020)

RichHTML

17

PDF

101

Abstract:

Geographical flows represent meaningful interactions of geographical objects between pairs of locations. Mining the spatio-temporal association patterns of geographical flows is of great significance for uncovering the spatio-temporal dependency and heterogeneity among flows, as well as understanding the underlying flow mechanisms and spatio-temporal interactions. Currently, there is an increasing number of methods for spatial association analysis of geographical flows. However, there is limited research considering the spatio-temporal coupling characteristics of geographical flows and focusing on the impact of time effects on association pattern detection. Accurately capturing the spatio-temporal dynamics of dependencies between geographical flows remains a challenging issue in the flow association analysis field. To address this gap, this paper extends the spatio-temporal cross K-function of point process to the flow spatio-temporal cross K-function. The method takes the geographical flow as the research object, which can be used to detect spatio-temporal association patterns between any two types of geographical flows. Specifically, the global flow spatio-temporal cross K-function can detect the overall association patterns of geographical flows in the study area, while the local flow spatio-temporal cross K-function can identify the spatio-temporal associations at the local scale that does not follow the global pattern. This work utilizes the flow spatio-temporal cross K-function to analyze the spatio-temporal association patterns between taxi flows and ride-hailing flows on Xiamen Island. The global results show an isolated pattern between taxi flows and ride-hailing flows, suggesting that there is no intense positive competition between the two types of vehicles on Xiamen island. Whereas, the local results indicate competition between the two types of vehicles during the morning, afternoon, and evening peak hours. This competition is mainly concentrated in several specific areas, such as residential areas to industrial parks, residential areas to the airport, train stations to business districts, and business districts to tourist attractions.

Key words: geographical flows, spatio-temporal association patterns, spatio-temporal cross K-function, ride-hailing service, taxi service

CLC Number: