An adapted geographically weighted LASSO (Ada-GWL) model for predicting subway ridership

Abstract

Ridership prediction at station level plays a critical role in subway transportation planning. Among various existing ridership prediction methods, direct demand model has been recognized as an effective approach. However, direct demand models including geographically weighted regression (GWR) have rarely been studied for local model selection in ridership prediction. In practice, acquiring insights into subway ridership under multiple influencing factors from a local perspective is important for passenger flow management and transportation planning operations adapting to local conditions. In this study, we propose an adapted geographically weighted LASSO (Ada-GWL) framework for modelling subway ridership, which involves regression-coefficient shrinkage and local model selection. It takes subway network layout into account and adopts network-based distance metric instead of Euclidean-based distance metric, making it so-called adapted to the context of subway networks. The real-world case of Shenzhen Metro is used to elaborate our proposed model. The results show that the proposed Ada-GWL model performs the best compared with the global model (ordinary least square, GWR, GWR calibrated with network-based distance metric and geographically weighted LASSO (GWL) in terms of estimation error and goodness-of-fit. Through understanding the variation of each coefficient across space (elasticities) and variables selection of each station, it provides more realistic conclusions based on local analysis. Besides, through clustering analysis of the stations according to the regression coefficients, clusters’ functional characteristics are found to be in compliance with the policy of functional land use in Shenzhen, indicating the high interpretability of Ada-GWL model from the spatial angle. In other words, the regression coefficients of different stations can provide us the local prospective to understand the influence of factors on stations’ ridership.

Appendix

Summary of literature review on direct demand models for ridership prediction

The related studies on direct demand models for ridership prediction are summarized as following (Figs. 15 and 16).

Fig. 15

Summary of literature review on direct demand models for ridership prediction

Fig. 16

Summary of literature review on direct demand models for ridership prediction (contd) (Cardozo et al. 2012; Gutiérrez et al. 2011; Choi et al. 2012; Cervero 2006; Chu 2004; Walters and Cervero 2003; Kuby et al. 2004; Sohn and Shim 2010; Loo et al. 2010; Sung and Oh 2011; Thompson et al. 2012; Guerra et al. 2012; Zhao et al. 2013; Chan and Miranda-Moreno 2013; Singhal et al. 2014; Liu et al. 2016; Pan et al. 2017; Jun et al. 2015; Zhang and Wang 2014; Taylor et al. 2003; Estupiñán and Rodríguez 2008; Deng and Xu 2015; Li et al. 2016; Hu et al. 2016)

Definitions of stations’ identifiers

For the sake of convenience in representing and understanding, we use alphanumeric code instead of Chinese to denote each station name. Here, we define identifiers for station names according to the following rules: (1) non-transfer stations consist of 3 digits, where the first digit denotes the line number, and the rest 2 digits denote the sequential number of station; (2) transfer stations start with character t followed by 3 digits, where the first 2 digits denote the intersection of two lines, and the last digit means the sequential number of intersections between those two lines. For example, “402” represents the 2nd station of Line 4, and “t131” represents the transfer station that is the first intersection of lines 1 and 3. In this way, all of 118 stations can be encoded by such identifiers containing the line and station information literally (Figs. 11 and 13).