Understanding Influencing Mechanisms of Subway Station-Line Peak Deviation Time: Spatial Econometric Perspective

Abstract

The peak deviation time (PDT) between subway stations and lines is a critical temporal misalignment in urban rail systems, yet it remains understudied. Accurately understanding PDT is essential for calibrating station-level peak-hour time (PHT) during planning, which directly informs station design ridership and facility configuration. Ignoring PDT can result in persistent supply–demand mismatches, exacerbating platform congestion and impairing operational efficiency. To address this gap, this study develops a spatial econometric framework to examine the spatiotemporal patterns and correlates of PDT. We implement a model selection procedure comparing the spatial lag model (SLM), spatial error model (SEM), and spatial Durbin model (SDM) with multiple network distance-based spatial weight matrices, enabling the characterization of spatial interdependencies among stations. Using multi-source data from the Xi’an subway system, the results, interpreted as conditional associations given the cross-sectional design, show that: (1) PDT exhibits significant spatiotemporal heterogeneity, with the SDM outperforming alternative models in capturing spatial autocorrelation; (2) the correlates of PDT are highly period- and direction-specific, with higher shares of recreational and medical land uses consistently associated with smaller temporal deviations across multiple peak periods, indicating their potential relevance as diagnostic indicators in station-area planning contexts; and (3) significant spatial spillover patterns are observed, with factors such as transport hub land ratio and terminal or transfer station attributes associated with both local and neighboring stations’ PDT. These findings underscore the importance of a network-wide perspective for station facility planning and operational management.