中文摘要：

为了提高综合交通枢纽等集散点的服务水平,针对依次发车和单独发车2种典型的泊位设置模式,以最小化出租车和乘客的综合平均排队时间为目标,建立了同时考虑泊位设置模式和规模的集成优化模型。基于Monte Carlo仿真算法模拟乘客和出租车的随机到达及交互过程,进行了模型求解;基于算例分析验证方法的可行性,给出了一定条件下泊位设置模式和规模的参照表,并分别针对依次发车和单独发车模式下的随机需求、乘客平均上车时间、乘客和出租车的到达分布进行了敏感性分析。结果表明：该方法综合考虑了乘客和出租车的排队状况,能给出随机需求下最优的上客区泊位设置模式和上客泊位数;且单独发车模式比依次发车模式具有更好的鲁棒性。

英文摘要：

In order to improve the service level of passenger distributing centers such as comprehensive transportation hubs,for two typical departure modes,namely orderly and separate modes,with the aim of minimizing overall average queuing time of taxis and passengers,an aggregated model was established with the management mode and scale of the pick-up zone simultaneously considered.The stochastic arrival and interaction process of taxis and passengers were simulated based on Monte Carlo algorithm,and the proposed model was solved.Based on the numerical example analysis,the method was validated.A table of the referential management mode and scale under certain conditions was suggested.A sensitivity analysis of stochastic demand of both modes,average boarding time,arrival distribution of taxis and passengers was conducted.The results show that the proposed method considers queuing conditions of both taxis and passengers.It yields the optimal management mode and scale of the pick-up zone under stochastic demand.The separate departure mode is more robust than the orderly one.