Mining Customized Bus Demand Spots Based on Smart Card Data: A Case Study of the Beijing Public Transit System

• Main Authors: Yiyi Cheng, Ailing Huang, Geqi Qi, Bei Zhang
• Format: Article
• Language: English
• Published: IEEE 2019-01-01
• Series: IEEE Access
• Subjects: Bus smart card data; customized bus; potential demand area; geographic information system; spatial clustering analysis
• Online Access: https://ieeexplore.ieee.org/document/8933013/

In recent years, to fix the shortcomings of traditional bus service and meet the diversified needs of passengers, a new type of transit system, the customized bus (CB), has been proposed. However, how to define and mine the CB's demand is still less being addressed. Since the data of bus smart cards can provide more travel information, it makes the mining of potential CB's demand spots more possible, which can be helpful in CB service design. In order to mine the demand spots more scientifically, this paper, for the first time, quantitatively defines the CB demand characteristics and criteria of selecting potential area, and develops a demand hotspots extraction methodology for CB. The methodology solves two issues primarily. One is how to organize massive smart card data and obtain the space-time pattern and mobility of passenger efficiently; the other is how to mix the CB demand characteristics into the method. This demand spots extraction method can generate multi-style maps, including the heat and origin-destination maps, for spatial cluster of CB's demand spots in rational areas in terms of the CB demand characteristics based on geographic information system. By using the bus smart card data in Beijing, China, this paper carries out a case study to validate the method. The empirical data mining analysis shows that our proposed method can define demand spots ideally. Our work can provide a valuable reference for decision makers to design CB system.