城市公交系统运营时段准确划分研究与仿真

Research and Simulation on Operation Period Division of Urban Bus System

公交车辆在全天不同时段的单程运行时间各不相同,对各时段公交线路运营规律对全天运营时段进行划分,并针对性地制定适应于不同时段的公交运营方案,是公交调度优化的重要基础。目前我国公交运营时段划分主要通过经验判断,主观性较强且难以反映公交运营规律。以研究公交线路不同时段发车的单程运行时间为属性变量,采用有序样本聚类中的Fisher算法,通过定义运营时段类别的类直径与分类损失函数,搜索分类损失函数最小状态下的最优运营时段划分结果。郑州市公共交通实例仿真结果表明,公交运营时段划分类数为9时,具有最好的分类效果。提出的基于有序样本聚类的公交运营时段方法有效可行,时段划分结果与实际更加吻合。

Great difference existed in the one-way running time of buses at variable periods such as peak and off-peak. It is an important foundation for the scheduling optimization to divide the whole day service time based on operation law of bus line in the different periods and to formulate the adaptable bus operation scheme to different time periods. The current bus operation time is mainly divided by artificial experience, which is too subjective to reflect the operation law of bus line. In order to get the optimal operation period division results, this article defined the class diameter and the classification loss function of the operation period category. Then it can be searched under the minimum state of the classified loss function, which taking the single run time of the bus departure time in different periods as the attribute variables and using the Fisher algorithm in the ordered sample clustering. The simulation of Zhengzhou public transportation shows that it is the best result with 9 operation periods. The proposed method in this paper based on ordered sample clustering is effective and feasible for dividing the operation period of buses, and the result of time division is more consistent with the actuality.