In this study, a stated preference (SP) survey based on a D-efficient experimental design was conducted for the Nanjing Metro to examine the passenger flow and elaborate the heterogeneity of the departure time choice behaviour of metro passengers during peak hours. A latent class logit model was applied to calibrate the data, and the attribute variables were studied using an elasticity analysis. The survey identified four classes of metro passengers in peak hours who have different preferences for the departure time choice. The results show that fixed working hours and arrival time constraints are the major attributes that determine the classification of the passengers, while travel costs and overcrowding are the key factors that affect the departure time choice. However, passengers are more sensitive to metro fares than crowdedness. By reducing travel costs and mitigating overcrowding, passengers’ choice of departure time, especially when deciding whether to depart earlier than usual, was significantly affected. This study aims to provide theoretical support and a basis for decision-making for metro operation departments to enact passenger flow control strategies.

中文翻译：

---

高峰时段地铁乘客出发时间选择分析

在这项研究中，针对南京地铁进行了基于D效率实验设计的陈述偏好（SP）调查，以检查客流并阐述高峰时段地铁乘客出发时间选择行为的异质性。应用潜在类logit模型校准数据，并使用弹性分析研究属性变量。该调查确定了四类在高峰时段对出发时间选择有不同偏好的地铁乘客。结果表明，固定工作时间和到达时间限制是决定乘客分类的主要属性，而旅行成本和拥挤是影响出发时间选择的关键因素。但是，乘客对地铁票价比拥挤更加敏感。通过降低旅行成本和减轻拥挤状况，特别是在决定是否比平时更早出发时，乘客选择出发时间的选择受到了极大的影响。本研究旨在为地铁运营部门制定客流控制策略提供理论支持和决策依据。