考虑时间窗的定制公交线路时空分层优化模型

摘要: 研究定制公交线网布局及调度优化对增强公交系统吸引力, 提高乘客出行效率具有重要意义。针对定制公交乘客需求点在时间和空间上分布离散的特点, 构建了考虑时间窗的定制公交时空分层优化模型, 并设计遗传算法对模型进行求解。通过渔网与核密度分析对需求点在时间和空间上进行了热点识别, 并实现热点区域聚类分析以及合乘站点分类。基于合乘站点集合, 综合考虑公交容量、线路长度、乘客出行距离构建了线路空间优化模型, 以乘客的时间花费最小作为优化目标构建了线路时间优化模型。以济南市城区定制公交为例对模型的性能进行评估, 案例结果表明: 模型优化后的线路方案, 乘客平均服务覆盖率可达96%, 服务区域内每个时段的单个乘客的平均节省时间为15 min, 公交的平均满载率为90%。
- 交通工程 /
- 定制公交 /
- 时空分层优化 /
- 空间聚类 /
- 遗传算法 /
- 时间窗
Abstract: Studying the layout and scheduling optimization of a customized bus line network has important implications, enhancing the attractiveness of the public transport system and passenger travel. However, the distribution of customized bus passengers' demand points in time and space is discrete, which hinders the bus line design. A timespace hierarchical optimization model of customized buses considering time windows is constructed to solve this problem, and a genetic algorithm is designed to solve the model. The hot spots of demand points are identified in time and space by analyzing the fishing net and kernel density, with the cluster analysis of hot spots and the classification of bus pooling realized. Based on the set of bus pooling, a space optimization model of bus lines is constructed by the bus capacity, line length, and passenger travel distance. A time optimization model of bus lines is constructed by the minimum time cost of passengers. Jinan customized bused are used to evaluate the performance of the model. The results show that the routing scheme is optimized by the model, with an average service coverage rate of passengers of 96%, the average travel time saved by the single passenger in each period of the service area of 15 minutes, and an average load factor of public transport of 90%.
- traffic engineering /
- customized bus /
- space-time hierarchical optimization /
- spatial clustering /
- genetic algorithm /
- time windows

HTML全文

图 1 算法流程

Figure 1. Algorithm flow

下载: 全尺寸图片幻灯片

图 2 需求点核密度

Figure 2. Kernel density of demand points

下载: 全尺寸图片幻灯片

图 3 上车需求渔网图

Figure 3. Hot spot map of boarding demand

下载: 全尺寸图片幻灯片

图 4 下车需求渔网图

Figure 4. Hot spot map of alighting demand

下载: 全尺寸图片幻灯片

图 5 站点分布图

Figure 5. Site distribution

下载: 全尺寸图片幻灯片

表 1 定制公交需求数据

Table 1. Demand data of customized buses

序号	云账户	上车点	下车点	上车时间	下车时间	返程时间	上车高德经度/(°)	上车高德纬度/(°)	下车高德经度/(°)	下车高德纬度/(°)
5575	account-180801	龙湖.春江郦城2期	达内济南万达中心	08:20	08:55	18:00	118.051 462	24.584 578	117.131 095	36.686 382
2905	account-274990	响泉路	山东和美华农牧科技股份有限公司	07:00	07:50	17:00	113.075 114	27.779 672	117.323 408	36.707 568
753	account-55545	加州小区24栋	龙奥公交枢纽(公交站)	07:15	08:15		115.568 31	28.842 056	117.137 306	36.654 095
5453	account-235766	文旅城(公交站)	济南水务集团有限公司(顺河东街)	07:30	08:15		106.349 068	29.634 964	117.008 424	36.665 363
3983	account-37676	翡翠华庭二期停车场(入口)	第四人民医院	07:00	07:30	17:15	106.511 279	29.643 769	116.978 9	36.694 529
845	account-32791	春江郦城	火车站(公交站)	07:44	17:55		121.582 684	29.984 336	116.992 531	36.667 751
3956	account-405	杭州市滨江区南环路1656号	杭州市江干区12号大街12~大街16号	08:30	18:30		120.200 954	30.177 261	120.388 441	30.289 529

下载: 导出CSV

表 2 需求分类统计

Table 2. Demand classification statistics

ID	上车站点	下车站点	上车时间	下车时间	下车时间	上车区域编号	下车区域编号
account-101566	1	14	06:15	07:30	07:35	2	1
account-121734	1	2	06:30	08:00	08:05	2	3
account-131791	1	4	06:30	07:40	07:45	2	3
account-137716	1	2	07:40	08:40	08:45	2	3
account-160457	1	8	07:30	08:30	08:35	2	1
account-169021	1	5	07:00	07:50	07:55	2	1
account-20811	1	2	07:00	08:00	08:05	2	3
account-20920	1	4	07:00	07:44	07:49	2	3

下载: 导出CSV

表 3 区域间不同时段需求人数

Table 3. Number of people needed in different periods between regions

时段	需求人数
1	12
2	44
3	46
4	14
5	0

下载: 导出CSV

表 4 普通算法求解的定制公交线路

Table 4. Customized bus routes solved by the common algorithm

编号	线路站点	线路里程/km	载客人数
1	57-82-1-114-117-115	40.28	12
2	57-1-82-114-117-119-122-115	63.08	39
3	95-5-114-115	27.75	5
4	5-1-95-97-114-117-115-122	44.36	46
5	57-1-5-114-115	35.84	14

下载: 导出CSV

表 5 本文算法求解的定制公交线路

Table 5. Customized bus route solved by the algorithm proposed in the work

时段	线路站点	线路里程/km	载客人数
06:00—06:30	82-57-1-114-117-115	39.36	12
06：30—07：00	1-114-117-119-122-115	30.68	37
07:00—07:30	5-97-1-95-114-117-115-122	44.94	46
07：30—08：00	57-1-5-114-115	35.16	14

下载: 导出CSV

表 6 定制公交线路方案

Table 6. Customized bus-route schemes

时段	线路站点线路里程/km	车型	载客人数	发车时间
06:00—06:30	82-57-1-114-117-115 39.36	小型	12	06:14
06:30—07:00	1-114-117-119-122-115 30.68	中型	37	06:57
07:00—07:30	95-1-5-97-114-117-115-122 35.24	大型	46	07:06
07:30—08:00	57-1-5-114-115 35.16	小型	14	07:30

下载: 导出CSV

表 7 线路方案评价表

Table 7. Evaluation of route schemes

时段	服务率/%	人均节省时间/min	满载率/%
06:00—06:30	100	34	80
06:30—07:00	84	7	93
07:00—07:30	100	5	92
07:30—08:00	100	16	93

下载: 导出CSV

参考文献(19)

[1]	HARMS S, TRUFFER B. The emergence of a nationwide carsharing coperative in Switzerland[R]. Holland: University of Twente, 1998.
[2]	TSUBOUCHI K, YAMATO H, HIEKATA K. Innovative on-demand bus system in Japan[J]. IET Intelligent Transport Systems, 2010, 4(4): 270-279. doi: 10.1049/iet-its.2009.0113
[3]	NEVEN A, BRAEKERS K, DECLERCQ K, et al. Assessing the impact of different policy decisions on the resource requirements of a demand responsive transport system for persons with Disabilities[J]. Transport Policy, 2015(44): 48-57. http://www.sciencedirect.com/science/article/pii/S0967070X15300305
[4]	FENG Q, LI W, JIAn Z. A dynamic station strategy to improve the performance of flex-route transit services[J]. Transportation Research Part C: Emerging Technologies, 2014, 48(12): 229-240. http://www.onacademic.com/detail/journal_1000036711989510_362a.html
[5]	CHANG S K, YU W J. Comparison of subsidized fixed and flexible-route bus systems[J]. Transportation Research Record, 1996(1557): 15-20.
[6]	郭戎格. 基于IC卡数据的定制公交线路优化[D]. 北京: 北京交通大学, 2017. GUO Rongge. Optimal routing design of customized bus based on bus ic data[D]. Beijing: Beijing Jiao Tong University, 2017. (in Chinese).
[7]	张鹤鹏. 基于GPS定位数据的出行行为研究[D]. 北京: 北京交通大学, 2019. ZHANG Hepeng. Study on the travel behavior based on GPSdata[D]. Beijing: Beijing Jiao Tong University, 2019. (in Chinese).
[8]	陈亚玲. 基于出租车GPS轨迹数据的公交路线优化[D]. 绵阳: 西南科技大学, 2019. CHEN Yaling. bus route optimization based on taxi gps trajectory data[D]. Mianyang: Southwest University of Science and Technology, 2019. (in Chinese).
[9]	曾勋. 基于IC卡和GPS数据的公交服务能力评估[D]. 西安: 长安大学, 2019. ZENG Xun. Bus service capacity assessm entbased on IC card and GPS data[D]. Xi'an: Chang'an University, 2019. (in Chinese).
[10]	杨敬锋, 张南峰, 杨骥, 等. 基于公交站群的定制公交出行需求热力图分析模型[C]. 第5届中国卫星导航与位置服务年会暨展览会, 成都: 中国卫星导航定位协会, 2016. YANG Jingfeng, ZHANG Nanfeng, Yang Ji, et al. Thermal graph analysis model of customized bus travel demand based on bus station group[C]. 5^thGNSS and LBS Annual Conference & amp; Exhibition, Chengdu: GNSS & amp; LBS Association of China, 2016. (in Chinese).
[11]	李莎. 定制公交线网的建模与智能优化算法研究[D]. 广州: 华南理工大学, 2017. LI Sha. A Dissertation submithed for the degree of master[D]. Guangzhou: South China University of Technology, 2017. (in Chinese).
[12]	靳文舟, 韩博文, 郝小妮, 等. 基于小波神经网络的定制公交目标乘客出行意愿预测[J]. 重庆交通大学学报(自然科学版), 2018, 37(8): 81-87. doi: 10.3969/j.issn.1674-0696.2018.08.14 JIN Wenzhou, HAN Bowen, HAO Xiaoni, et al. Prediction of target passenger'travel intention of customized public transport based on wavelet neural network[J]. Journal of Chongqing Jiaotong University(Natural Science), 2018, 37(8): 81-87. (in Chinese). doi: 10.3969/j.issn.1674-0696.2018.08.14
[13]	LI Wenyong, YU Ziwei, WEI Shuo, et al. Custom bus optimization of docking stations based on partition path selection[J]. Advances in Mechanical Engineering, 2017, 9(10): 1-6. http://www.onacademic.com/detail/journal_1000040145143610_11c2.html
[14]	雷永巍, 林培群, 姚凯斌. 互联网定制公交的网络调度模型及其求解算法[J]. 交通运输系统工程与信息, 2017, 17(1): 157-163. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201701024.htm LEI Yongwei, LIN Peiqun, YAO Kaibin. The network scheduling model and its solution algorithm of internet customized shuttle bus[J]. Journal of Transportation Systems Engineering and Information Technology, 2017, 17(1): 157-163. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT201701024.htm
[15]	李颖. 定制公交的决策与算法研究[D]. 南京: 南京林业大学, 2017. LI Ying. Research on custom public transport decision and algorithm[D]. Nanjing: Nanjing Forestry University, 2017. (in Chinese).
[16]	陶浪, 马昌喜, 朱昌锋, 等. 基于遗传算法的定制公交路线多目标优化[J]. 兰州交通大学学报, 2018, 37(2): 31-37. doi: 10.3969/j.issn.1001-4373.2018.02.006 TAO Lang, MA Changxi, ZHU Changfeng, et al. Multiobjective optimization of customized bus routes based on genetic algorithm[J]. Journal of Lanzhou Jiaotong University, 2018, 37(2): 31-37. (in Chinese). doi: 10.3969/j.issn.1001-4373.2018.02.006
[17]	王健, 曹阳, 王运豪. 考虑出行时间窗的定制公交线路车辆调度方法[J]. 中国公路学报, 2018, 31(5): 143-150. doi: 10.3969/j.issn.1001-7372.2018.05.017 WANG Jian, CAO Yang, WANG Yunhao. Customized bus route vehicle schedule method considering travel time windows[J]. China Journal of Highway and Transport, 2018, 31(5): 143-150. (in Chinese). doi: 10.3969/j.issn.1001-7372.2018.05.017
[18]	张宇宁, 武建平. 面向通勤需求的定制公交线路规划研究[J]. 黑龙江交通科技, 2019, 42(12): 147-148. doi: 10.3969/j.issn.1008-3383.2019.12.078 ZHANG Yuning, WU Jianping. Research on customized bus route planning for commuting demand[J]. Communications Science and Technology Heilongjiang, 2019, 42(12): 147-148. (in Chinese). doi: 10.3969/j.issn.1008-3383.2019.12.078
[19]	陈汐, 王印海, 刘剑锋, 等. 多区域通勤定制公交线路规划模型及求解算法[J]. 交通运输系统工程与信息, 2020, 20(4): 166-172+186. https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202004024.htm CHEN Xi, WANG Yinhai, Liu Jianfeng, et al. Multi-region commuting bus route design model and solution method[J]. Journal of Transportation Systems Engineering and Information Technology, 2020, 20(4): 166-172+186. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-YSXT202004024.htm