Causal effect of metro operation on regional resident mobility considering zone-based trip time reliability

The full use of underground space is recognized as a necessary means to improve the accessibility of residents and change the spatial distribution of urban population. The dividends generated by the metro operation will increase as the increases in the number of residents attracted to the zone. However, when the number of residents within the zone exceeds a certain threshold, travel accessibility may be reduced by exceeding the optimal service capacity of the metro, which leads to a decline in the per capita dividend. This paper uses the zone-based trip time reliability (TTR) to represent travel accessibility, and describes the interaction between the zone-based TTR and regional resident mobility. A generalized boosted method based Rubin causal model with longitudinal data was adopted to investigate the impacts of the improvement of zone-based TTR by metro operation on regional resident mobility. Then, an approximate BPR function was established to represent the feedback impact of regional resident mobility on zone-based TTR. Five typical areas considering population, average age, income level and land value are proposed to recognize the resident spatial re-segmentation rule. The results show that the metro operation has a significant impact on the regional resident mobility. And the improvement of zone-based TTR by metro operation is more attractive for resident relocation in Mixed and Stable areas than in Vibrant and Siphonic areas under the similar location distance. This paper intends to indicate the trade-off between zone-based trip time reliability (TTR) and regional resident mobility, and provide a reference for the development and utilization of urban underground space.