基于多尺度时空聚类的共享单车潮汐特征挖掘与需求预测研究

doi:10.12082/dqxxkx.2022.210691

摘要/Abstract

摘要：

当前,我国政府和单车企业多以划定电子围栏停车点的方式进行共享单车的规范化管理,由于单个电子围栏内部单车流入流出的随机性和不确定性较大,以单个围栏为单位进行单车管理的工作量大且不具现实意义。因此,有必要对电子围栏停车点进行聚类划分,实行区域化的管理与调度。基于此,本文提出一种基于时空约束的网络图聚类算法,该算法综合考虑空间因素（地理位置、地理环境特征）和时间因素（历史订单）,只需通过距离阈值设定即可实现电子围栏的多尺度聚类划分,实验分别在3000 m和700 m距离阈值条件下对厦门岛和乌石浦地区电子围栏进行聚类,结果显示该算法不仅能够将具有相似时空特征的电子围栏聚到同一社区簇内,而且能够使得单车流动主要集中在划分后的社区内部;随后,在社区划分基础上进行单车潮汐特征挖掘,能够有效识别和定位单车使用的热点地区;最后,利用长短时记忆神经网络（Long-Short Time Memory network, LSTM）进行单车订单需求预测,结果显示有84%以上社区的预测准确率在85%以上,平均预测准确率为91.301%,预测效果较好,可有效满足单车调度需求。本文研究成果可服务于电子围栏停车点规划与共享单车的区域化管理与调度工作。

关键词: 共享单车, 电子围栏, 数据挖掘, 时空约束, 多尺度聚类, 使用模式, 需求预测, 厦门岛

Abstract:

At present, China government and bike-sharing companies mostly use electronic fence parking stations to manage the shared bicycles normatively. Electric fence parking stations for free-floating bike-sharing are predetermined 'virtual fences' to guide users to park bikes in designated zones and regulate inappropriate parking behaviors. However, due to the randomness and uncertainty of the inflow and outflow of bicycles at a single parking station, the scheduling of bicycles based on an independent parking station is hard to realize. Therefore, it is necessary to group fence stations into clusters and implement regional management. In this paper, we proposed a network clustering algorithm based on spatiotemporal constraints, which comprehensively considered spatial factors (location and geographical environment of the parking stations) and temporal factors (historical bike-sharing system orders) as the clustering partition basis, and this algorithm can realize the multi-scale groups division of parking stations only by setting a distance threshold. We chose Xiamen Island as the research region. Using the distance thresholds of 3000 m and 700 m respectively, we carried out clustering experiments on the electronic fence parking stations in the whole Xiamen Island and its Wushipu block. The results showed that this algorithm can not only gather the parking stations with similar temporal and spatial characteristics into the same group, but also make the shared bike flow mainly concentrated in the streets within each group, which is convenient for regional management. Then, we mined the characteristics of shared bikes among the partitioned groups, which can effectively identify and locate hot areas for shared bikes. The results showed that subway stations, office buildings, parks, hospitals, shopping malls, and residential areas had a greater impact on the usage pattern of shared bikes. In particular, it is necessary to focus on the accumulation of shared bikes near office buildings, shopping malls, hospitals, and subway stations, and the shortage of bicycles near the residential areas, parks, and factories during the morning rush hours. Finally, we used the Long Short Time Memory network (LSTM) to predict the orders of shared bikes. The results showed that 84% of the groups had a prediction accuracy of more than 85%, and the average of the overall prediction accuracy was 91.301%, which can meet the needs of bike-sharing system scheduling. Our research provides scientific suggestions for relevant departments to arrange electronic fence parking stations, and the LSTM model has high accuracy in predicting bicycle flow, which is effective in reducing the scheduling cost of bike-sharing system and improve the management efficiency.

Key words: bike-sharing system, electronic fence, data mining, spatiotemporal constraints, multiscale clustering, usage pattern, demand forecast, Xiamen island

姜晓, 白璐斌, 楼夏寅, 李梅, 刘晖. 基于多尺度时空聚类的共享单车潮汐特征挖掘与需求预测研究[J]. 地球信息科学学报, 2022, 24(6): 1047-1060.DOI:10.12082/dqxxkx.2022.210691

JIANG Xiao, BAI Lubin, LOU Xiayin, LI Mei, LIU Hui. Usage Patterns Identification and Flow Prediction of Bike-sharing System based on Multiscale Spatiotemporal Clustering[J]. Journal of Geo-information Science, 2022, 24(6): 1047-1060.DOI:10.12082/dqxxkx.2022.210691

图/表 18

图1

表1

图2

图3

图4

图5

算法1 基于时空约束的网络聚类GC²
Require:
输入:停车点相关性网络矩阵,G;
节点集合,V
一次迭代中节点类别交换的最低次数, $C m i n$
Ensure: 给每个节点初始化一个独一无二的簇标签
repeat 初始化交换次数 $C c$ =0 for $v i$ in V do 从节点集合V中移除当前节点 $v i,$ 记录 $v i$ 此时的标签 $l a b e l b e f o r e$ 计算节点 $v i$ 与其邻接簇之间的收益value
将节点 $v i$ 分配给value最大的簇,将此簇的标签 $l a b e l a f t e r$ 赋予 $v i$ ,将 $v i$ 添加到集合V中
if $l a b e l b e f o r e ≠ l a b e l a f t e r$ then
$C c = C c + 1$ end if end if until $C c ≤ C m i n$

图6

图7

图8

图9

图10

图11

表3

表4

图12

表5

图13

参考文献 33

[1]	邵鹏, 王齐, 赵超. 共享单车绿色使用行为与意愿的影响因素研究[J]. 干旱区资源与环境, 2020, 34(3):64-68.
	[ Shao P, Wang Q, Zhao C. Research on the factors influencing shared bicycle green use behavior and intention[J]. Journal of Arid Land Resources and Environment, 2020, 34(3):64-68. ] DOI: 10.13448/j.cnki.jalre.2020.67 doi: 10.13448/j.cnki.jalre.2020.67
[2]	Zhang Y P, Lin D, Mi Z F. Electric fence planning for dockless bike-sharing services[J]. Journal of Cleaner Production, 2019, 206:383-393. DOI: 10.1016/j.jclepro.2018.09.215 doi: 10.1016/j.jclepro.2018.09.215
[3]	Singla A, Santoni M, Bartok G, et al. Incentivizing users for balancing bike sharing systems[C]. Proceedings of the Twenty-Ninth Aaai Conference on Artificial Intelligence, 2015.
[4]	Rokach L, Maimon O. Clustering methods[M]. Boston: Springer US. 2005. DOI: 10.1007/0-387-25465-X_15 doi: 10.1007/0-387-25465-X_15
[5]	章永来, 周耀鉴. 聚类算法综述[J]. 计算机应用, 2019, 39(7):1869-1882.
	[ Zhang Y L, Zhou Y J. Review of clustering algorithms[J]. Journal of Computer Applications, 2019, 39(7):1869-1882. ] DOI: 10.11772/j.issn.1001-9081.2019010174 doi: 10.11772/j.issn.1001-9081.2019010174
[6]	高楹, 宋辞, 郭思慧,等. 接驳地铁站的共享单车源汇时空特征及其影响因素[J]. 地球信息科学学报, 2021, 23(1):155-170. doi: 10.12082/dqxxkx.2021.200351
	[ Gao Y, Song C, Guo S H, et al. Spatial-temporal characteristics and influencing factors of source and sink of dockless sharing bicycles connected to subway stations[J]. Journal of Geo-information Science, 2021, 23(1):155-170. ] DOI: 10.12082/dqxxkx.2021.200351 doi: 10.12082/dqxxkx.2021.200351
[7]	靳爽, 庞明宝. 基于K-means的城市轨道交通社区接驳共享单车停靠点规划[J]. 科学技术与工程, 2019, 19(30):343-347.
	[ Jin S, Pang M B. Planning of shared bicycle stop for urban rail transit community connection based on K-means[J]. Science Technology and Engineering, 2019, 19(30):343-347. ]
[8]	Yu J J, Ji Y J, Yi C Y, et al. Estimating model for urban carrying capacity on bike-sharing[J]. Journal of Central South University, 2021, 28(6):1775-1785. DOI: 10.1007/s11771-021-4661-6 doi: 10.1007/s11771-021-4661-6
[9]	Hua M Z, Chen X W, Zheng S J, et al. Estimating the parking demand of free-floating bike sharing: A journey-data-based study of Nanjing, China[J]. Journal of Cleaner Production, 2020, 244:1-11. DOI: 10.1016/j.jclepro.2019.118764 doi: 10.1016/j.jclepro.2019.118764
[10]	Jia W Z, Tan Y Y, Liu L, et al. Hierarchical prediction based on two-level Gaussian mixture model clustering for bike-sharing system[J]. Knowledge-Based Systems, 2019, 178:84-97. DOI: 10.1016/j.knosys.2019.04.020 doi: 10.1016/j.knosys.2019.04.020
[11]	Du Y C, Deng F W, Liao F X. A model framework for discovering the spatio-temporal usage patterns of public free-floating bike-sharing system[J]. Transportation Research Part C-Emerging Technologies, 2019, 103:39-55. DOI: 1 0.1016/j.trc.2019.04.006 doi: 1 0.1016/j.trc.2019.04.006
[12]	刘畅. 共享单车需求预测及调度研究[D]. 武汉: 武汉理工大学, 2018.
	[ Liu C. Research on the demand forecast and scheduling of bike-sharing[D]. Wuhan: Wuhan University of Technology, 2018. ]
[13]	Jia W Z, Tan Y Y, Li J. Hierarchical prediction based on two-level affinity propagation clustering for bike-sharing system[J]. IEEE Access, 2018, 6:45875-45885. DOI: 10.1 109/ACCESS.2018.2865658 doi: 10.1 109/ACCESS.2018.2865658
[14]	Ciancia V, Latella D, Massink M, et al. Exploring spatio-temporal properties of bike-sharing systems[C]. 2015 IEEE International Conference on Self-Adaptive and Self-Organizing Systems Workshops (SASOW). IEEE, 2015. DOI: 10.1109/SASOW.2015.17
[15]	数字中国建设峰会. 2021数字中国创新大赛之大数据赛道-城市管理大数据专题[DB/OL].(2021-1-25)[2021-1-31].https://dcic.datafountain.cn/competitions/10 015 .
	[ Digital China Summit.Digital China innovation contest-the big data of urban management, DCIC 2021. [DB/OL]. (2021-1-28)[2021-1-28].https://dcic.datafountain.cn/competitions/10 015 . ]
[16]	高德地图. 厦门岛地区POI数据与城市路网数据[DB/OL].(2021-1-31)[2021-1-31]. https://ditu.amap.com/ .
	[ AmapThe POI data and urban road network data of Xiamen island[DB/OL]. (2021-1-31)[2021-1-31]. https://ditu.amap.com/ . ]
[17]	张景奇, 史文宝, 修春亮. POI数据在中国城市研究中的应用[J]. 地理科学, 2021, 41(1):140-148. doi: 10.13249/j.cnki.sgs.2021.01.015
	[ Zhang J Q, Shi W B, Xiu C L. Urban research using points of interest data in China[J]. Scientia Geographica Sinica, 2021, 41(1):140-148. ] DOI: 10.13249/j.cnki.sgs.2021.01.015 doi: 10.13249/j.cnki.sgs.2021.01.015
[18]	Xing Y Y, Wang K, Lu J J. Exploring travel patterns and trip purposes of dockless bike-sharing by analyzing massive bike-sharing data in Shanghai, China[J]. Journal of Transport Geography, 2020, 87:1-15. DOI: 10.1016/j.jtrangeo.2020.102787 doi: 10.1016/j.jtrangeo.2020.102787
[19]	Wang L Z, Bao X G, Chen H M, et al. Effective lossless condensed representation and discovery of spatial co-location patterns[J]. Information Sciences, 2018, 436:197-213. DOI: 10.1016/j.ins.2018.01.011 doi: 10.1016/j.ins.2018.01.011
[20]	Alcorn L G, Jiao J. Bike-sharing station usage and the surrounding built environments in major Texas cities[J]. Journal of Planning Education and Research, 2019, 3:1-14. DOI: 10.1177/0739456X19862854 doi: 10.1177/0739456X19862854
[21]	禹文豪, 艾廷华. 核密度估计法支持下的网络空间POI点可视化与分析[J]. 测绘学报, 2015, 44(1):82-90.
	[ Yu Y H, AI T H, The visualization and analysis of POI features under network space supported by kernel density estimation[J]. Acta Geodaetica et Cartographica Sinica, 2015, 44(1):82-90 ] DOI: 10.11947/j.AGCS.2015.20130538 doi: 10.11947/j.AGCS.2015.20130538
[22]	Zhang S C. Cost-sensitive KNN classification[J]. Neurocomputing, 2020, 391:234-242. DOI: 10.1016/j.neucom.2018.11.101 doi: 10.1016/j.neucom.2018.11.101
[23]	Kapuku C, Kho S Y, Kim D K, et al. Modeling the competitiveness of a bike-sharing system using bicycle GPS and transit smartcard data[J]. Transportation Letters-the International Journal of Transportation Research, 2020,1-5. DOI: 10.1080/19427867.2020.1758389 doi: 10.1080/19427867.2020.1758389
[24]	Fortunato S. Community detection in graphs[J]. Physics Reports, 2010, 486(3-5):75-174. DOI: 10.1016/j.physrep.2009.11.002 doi: 10.1016/j.physrep.2009.11.002
[25]	Liu X, Murata T. Advanced modularity-specialized label propagation algorithm for detecting communities in networks[J]. Physica a-Statistical Mechanics and Its Applications, 2010, 389(7):1493-1500. DOI: 10.1016/j.physa.2009.12.019 doi: 10.1016/j.physa.2009.12.019
[26]	Chen L B, Zhang D Q, Wang L Y, et al. Dynamic cluster-based over-demand prediction in bike sharing systems[C]. Ubicomp'16: Proceedings of the 2016 Acm International Joint Conference on Pervasive and Ubiquitous Computing, 2016. DOI: 10.1145/2971648.2971652
[27]	Sherstinsky A. Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) network[J]. Physica D: Nonlinear Phenomena, 2020, 404(8):132306. DOI: 10.1016/j.physd.2019.132306 doi: 10.1016/j.physd.2019.132306
[28]	Zhang C, Zhang L N, Liu Y D, et al. Short-term prediction of bike-sharing usage considering public transport: a LSTM approach[C]. IEEE International Conference on Intelligent Transportation Systems-ITSC, 2018. DOI: 10. 109/ITSC.2018.8569726 doi: 10. 109/ITSC.2018.8569726
[29]	付俐哲. 基于时空聚类与LSTM神经网络的共享单车需求预测模型[D]. 兰州: 西北师范大学, 2021.
	[ Fu L Z, Spatiotemporal clustering and LSTM based prediction model of bicycle sharing[D]. Lanzhou: Northwest Normal University, 2021 ]DOI: 10.27410/d.cnki.gxbfu.2021.001767 doi: 10.27410/d.cnki.gxbfu.2021.001767
[30]	曹旦旦, 范书瑞, 张艳,等. 基于长短期记忆神经网络模型的共享单车短时需求量预测[J]. 科学技术与工程, 2020, 20(20):8344-8349.
	[ Cao D D, Fan S R, Zhang Y, et al. Short-term demand forecasting of shared bicycles based on long short-term memory neural net-work model[J]. Science Technology and Engineering, 2020, 20(20):8344-8349. ]
[31]	Wang W J, Lu Y M. Analysis of the mean absolute error (MAE) and the root mean square error (RMSE) in assessing rounding model[C]. International Conference on Mechanical Engineering, 2018, 324(1). DOI: 10.1088/1757-899X/324/1/012049 doi: 10.1088/1757-899X/324/1/012049
[32]	万敏. 基于数据的共享单车需求预测和调度研究[D]. 南京: 南京大学, 2020.
	[ Wan M, Research on forecasting and scheduling of shared bicycle demand based on data[D]. Nanjing: Nanjing University, 2020. ] DOI: 10.27235/d.cnki.gnjiu.2020.000120 doi: 10.27235/d.cnki.gnjiu.2020.000120
[33]	Adler J, Parmryd I. Quantifying colocalization by correlation: The pearson correlation coefficient is superior to the mander's overlap coefficient[J]. Cytometry Part A, 2010, 77a(8):733-742. DOI: 10.1002/cyto.a.20896 doi: 10.1002/cyto.a.20896

数据名称	数据时间	数据规模	数据描述
数据名称	数据时间	数据规模	字段名称	字段含义
厦门岛共享单车订单数据	2020年12月21—25日6：00 am—10：00 am	58万条左右	BICYCLE_ID	加密后的单车ID号
			LATITDUE	纬度/°
			LONGITUDE	经度/°
			LOCK_STATUS	锁状态
			UPDATE_TIME	锁状态更新的时间
厦门岛共享单车电子围栏数据	2020年12月	1.4万个左右	FENCE_ID	电子围栏唯一编号
厦门岛共享单车电子围栏数据	2020年12月	1.4万个左右	FENCE_LOC	电子围栏位置坐标串
厦门岛POI数据	2021年1月	8000条左右	POI_TYPE	POI地物类别
			LATITDUE	纬度/°
			LONGITUDE	经度/°
厦门岛路网数据	2021年1月	8000条道路	Length	道路长度/m
厦门岛路网数据	2021年1月	8000条道路	name	道路名称

类别	数目	工厂	医院	公园	学校	地铁	办公楼	公交	居民区	商场	指数和
高频	8	0.262	0.333	0.480	0.214	0.544	0.656	0.078	0.318	0.323	3.209
中频	9	0.212	0.505	0.544	0.302	0.544	0.497	0.201	0.197	0.280	3.281
低频	8	0.204	0.153	0.247	0.105	0.178	0.300	0.054	0.275	0.098	1.613

类别	数目	工厂	医院	公园	学校	地铁	办公楼	公交	居民区	商场	指数和
流入	26	0.210	0.526	0.385	0.365	0.654	0.467	0.567	0.208	0.441	3.822
流出	34	0.221	0.382	0.407	0.355	0.548	0.323	0.548	0.222	0.298	3.302

社区	评价指标
社区	MAE	RMSE	PEARSON/%	AcR/%
0	12.394	26.682	84.651	86.888
1	27.065	37.996	97.572	94.228
2	21.540	38.921	96.790	96.674
3	60.711	96.166	92.481	86.226
4	48.237	66.411	98.163	97.235
5	11.158	16.725	98.011	95.098
6	6.461	8.509	97.987	95.845
7	33.671	42.844	99.099	95.559
8	5.448	10.145	75.370	91.557
9	44.250	64.302	97.784	97.264
10	25.329	55.762	98.099	95.295
11	19.842	29.365	99.111	97.393
12	35.355	50.229	99.335	95.543
13	2.250	3.806	54.333	78.498
14	7.0785	9.752	95.637	93.854
15	7.211	10.692	92.511	93.463
16	9.106	17.276	78.129	89.304
17	9.171	19.647	87.105	92.711
18	59.250	108.202	95.537	87.758
19	45.013	70.014	97.275	96.156
20	15.644	29.260	98.862	97.189
21	33.316	47.215	99.282	95.957
22	49.750	79.096	90.956	76.763
23	11.092	22.382	68.255	83.070
24	1.671	2.315	82.898	73.008
均值	24.080	38.548	91.010	91.301