A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Spatiotemporal Evolution of Travel Pattern Using Smart Card Data
Automated fare collection (AFC) systems can provide tap-in and tap-out records of passengers, allowing us to conduct a comprehensive analysis of spatiotemporal patterns for urban mobility. These temporal and spatial patterns, especially those observed over long periods, provide a better understanding of urban transportation planning and community historical development. In this paper, we explored spatiotemporal evolution of travel patterns using the smart card data of subway traveling from 2011 to 2017 in Shenzhen. To this end, a Gaussian mixture model with expectation–maximization (EM) algorithm clusters the travel patterns according to the frequency characteristics of passengers’ trips. In particular, we proposed the Pareto principle to negotiate diversified evaluation criteria on model parameters. Seven typical travel patterns are obtained using the proposed algorithm. Our findings highlighted that the proportion of each pattern remains relatively stable from 2011 to 2017, but the regular commuting passengers play an increasingly important position in the passenger flow. Additionally, focusing on the busiest commuting passengers, we depicted the spatial variations over years and identified the characters in different periods. Their cross-year usage of smart cards was finally examined to understand the migration of travel patterns over years. With reference to these methods and insights, transportation planners and policymakers can intuitively understand the historical variations of passengers’ travel patterns, which lays the foundation for improving the service of the subway system.
Spatiotemporal Evolution of Travel Pattern Using Smart Card Data
Automated fare collection (AFC) systems can provide tap-in and tap-out records of passengers, allowing us to conduct a comprehensive analysis of spatiotemporal patterns for urban mobility. These temporal and spatial patterns, especially those observed over long periods, provide a better understanding of urban transportation planning and community historical development. In this paper, we explored spatiotemporal evolution of travel patterns using the smart card data of subway traveling from 2011 to 2017 in Shenzhen. To this end, a Gaussian mixture model with expectation–maximization (EM) algorithm clusters the travel patterns according to the frequency characteristics of passengers’ trips. In particular, we proposed the Pareto principle to negotiate diversified evaluation criteria on model parameters. Seven typical travel patterns are obtained using the proposed algorithm. Our findings highlighted that the proportion of each pattern remains relatively stable from 2011 to 2017, but the regular commuting passengers play an increasingly important position in the passenger flow. Additionally, focusing on the busiest commuting passengers, we depicted the spatial variations over years and identified the characters in different periods. Their cross-year usage of smart cards was finally examined to understand the migration of travel patterns over years. With reference to these methods and insights, transportation planners and policymakers can intuitively understand the historical variations of passengers’ travel patterns, which lays the foundation for improving the service of the subway system.
Spatiotemporal Evolution of Travel Pattern Using Smart Card Data
Mu Lin (author) / Zhengdong Huang (author) / Tianhong Zhao (author) / Ying Zhang (author) / Heyi Wei (author)
2022
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
Travel behavior analysis using smart card data
| Springer Verlag | 2015
Travel behavior analysis using smart card data
| Online Contents | 2016
Probabilistic model for destination inference and travel pattern mining from smart card data
| Online Contents | 2020