考虑通勤者出行与活动之间的关联,建立了基于活动方法的出发时间选择模型,分析了常数和线性边际活动效用下地铁通勤者早高峰上班出发时间决策问题,并得出了模型的均衡性质:常数边际活动效用下的出发率是出发时间的线性函数,而线性边际活动效用下的出发率则是非线性函数.最后,算例验证了理论模型的结论,展示了出发率与相关参数之间的关系,还将基于活动方法的均衡结果与基于出行的均衡结果进行比较.结果表明:基于出行的方法会高估最早出发时间和最晚出发时间,低估出行区间长度,并高估最高出发率.当车内拥挤系数、发车间隔和总通勤人数分别增大时,在最早出发时刻,常数边际活动效用下的通勤者比线性边际活动效用下的通勤者更为敏感,且两种活动效用下的通勤者在最早出发时刻比在最晚出发时刻更为敏感.

Considering the association between commuter's activity schedule and trip schedule, this paper proposed a departure time choice model by introducing an activity-based approach. The model analyzes the transit commuters' departure time decision making problem in the morning peak with constant and linear marginal activity utility functions, and examines the equilibrium properties of the model: The departure rate curve is linear with constant marginal activity utility, and the departure rate curve is non-linear with linear marginal activity utility. Finally, a numerical example is conducted to verify the findings in the theoretical part and reveal the relationship between departure rate and related parameters. We also compare the departure time choices between the activity-based approach and the trip-based approach. The results show that the trip-based approach always overestimates the earliest departure time and latest departure time, underestimates the length of the departure interval, and overestimates the highest departure rate. When the parameters such as the in-carriage congestion coefficient, departure headway and total number of commuters increase respectively, commuters with constant marginal activity utility function are more sensitive than those with linear marginal activity utility function, and the commuters at the earliest departure time are more sensitive than those at the latest departure time with both kinds of activity utility functions.