论文总字数：49695字

摘 要

随着城市化进程的不断加快，机动车保有量迅速增长，城市交通问题，以及由此衍生的能源问题以及环境问题日益显著。随着慢行交通理念的逐渐推广，公共自行车作为一种便捷、绿色、经济的短距离出行方式，是地面常规公交以及城市轨道交通的重要补充，其在完善交通接驳、提高出行效率、改善出行体验等方面具有重要作用。通勤出行是城市居民出行中最基本也最重要的组成部分，探究公共自行车用户的通勤出行特征，准确把握公共自行车系统在通勤出行中的功能定位，对于引导公共自行车系统进一步发展具有重要的实际意义。

本文以南京市公共自行车系统为研究对象，对公共自行车用户通勤出行特征进行了研究。首先结合南京市公共自行车发展状况以及公共自行车通勤特性，对南京公共自行车两种主要的通勤模式：“直达型”通勤以及“轨道交通接驳型”通勤进行了初步分析。

其次，在掌握POI（Points of Interest，简称POI）、公共自行车站点信息，公共自行车站点刷卡数据等多源数据的基础上，进行了公共自行车通勤出行识别方法设计，使用两种聚类算法对站点类型进行了聚类分析；同时以南京市主城区为应用案例，依据租赁者的完整出行链识别出“直达型”和“轨道交通接驳型”通勤出行，并对识别结果进行了初步分析以及验证，结果显示识别方法具有一定的实践意义。

最后，结合数据挖掘以及可视化方法，从时空角度对比分析了两种通勤出行模式下各方面特征的异同点以及共享单车对南京市公共自行车的影响，结果显示“直达型”通勤的平均出行时长、平均出行距离均大于“轨道交通接驳型”通勤，两者在用户结构以及空间分布上也具有差异性；另一方面，共享单车对于公共自行车的影响主要体现于用户组成以及租赁总量的改变上，且“直达型”通勤出行受共享单车影响程度较低。

关键词：公共自行车，通勤识别，聚类分析，数据挖掘，出行特征

Abstract

With the accelerating process of urbanization, the rapid growth of motor vehicle ownership, urban transportation problems, and the resulting energy issues and environmental issues are increasingly significant. With the gradual promotion of the concept of slow traffic, public bicycles as a convenient, green, and economical short-distance travel mode are important supplements to regular ground transportation and urban rail transit. They improve traffic connections, improve travel efficiency, and improve Travel experience and other aspects have an important role. Commuter travel is the most basic and important component of urban residents' trips. Exploring the commuting trip characteristics of public bicycle users and accurately grasping the functional positioning of the public bicycle system in commuting travel have important practical significance for guiding the further development of public bicycle systems. .

This paper takes Nanjing public bicycle system as the research object, and studies the commuting trip characteristics of public bicycle users. Firstly, based on the development status of public bicycles in Nanjing and the characteristics of public bicycle commuting, the two main commuting modes of Nanjing public bicycles, namely “direct” commutes and “rail transport connection” commutes, were analyzed.

Secondly, based on multi-source data such as POI (Points of Interest, POI), public bicycle site information, and public bicycle site swiping data, a public bike commuter travel identification method was designed, and two clustering algorithms were used on the site. Types were clustered; at the same time, the main city of Nanjing was used as an application case to identify “direct” and “rail transport type” commuter travel based on the complete traveler’s travel chain, and conducted a preliminary analysis of the recognition results. Verification, the results show that the identification method has a certain practical significance.

Finally, combined with data mining and visualization methods, the similarities and differences in the characteristics of various aspects under the two commuting trip modes are compared and analyzed from the perspective of time and space, and the effects of shared bicycles on Nanjing public bicycles are shown. The results show that the average travel time of “connected” commutes. The average travel distance and travel volume are less than that of the “rail transport type” commuter. The two also have differences in user structure and spatial distribution. On the other hand, the impact of shared bicycles on public bicycles is mainly reflected in the user’s composition and lease. The change in the total amount, and the "direct" commuter travel is less affected by the sharing of bicycles.

KEY WORDS: Public Bicycle, Commuting Behavior, Data Mining, Clustering Analysis, Traveling Spatio-temporal

目录

摘要 I

Abstract II

第一章 绪论 1

1.1 研究背景与意义 1

1.1.1 研究背景 1

1.1.2 研究意义 2

1.2 国内外研究现状 3

1.2.1 国内外研究现状 3

1.2.2 现有研究不足 5

1.2.3 南京市公共自行车系统功能定位 5

1.3 研究内容与技术路线 6

1.3.1 研究内容 6

1.3.2 技术路线 6

第二章 基于刷卡数据和POI数据的公共自行车通勤出行识别方法 8

2.1 公共自行车通勤出行模式分析 8

2.1.1 “直达型”通勤 8

2.1.2 “轨道交通接驳型”通勤 8

2.2 公共自行车通勤出行识别方法设计 8

2.3 公共自行车站点聚类分析方法 9

2.3.1 K-Means聚类算法 10

2.3.2 DBSCAN聚类算法 10

2.4 “轨道交通接驳型”通勤出行识别方法 11

2.5 “直达型”通勤出行识别方法 11

2.6 本章小结 12

第三章 南京市公共自行车通勤出行识别分析 13

3.1 站点信息数据以及POI数据结构介绍及预处理 13

3.1.1 站点信息数据 13

3.1.2 POI数据 13

3.2 站点聚类结果分析 15

3.2.1 K-Means聚类结果分析 15

3.2.2 DBSCAN聚类结果分析 17

3.2.3 通勤识别站点类型限定 17

3.3 公共自行车通勤出行识别 17

3.3.1 刷卡数据结构介绍 17

3.3.2 数据清洗 18

3.3.3 数据预处理 18

3.4 公共自行车通勤出行识别结果分析 19

3.4.1 通勤出行识别结果分析 19

3.5 本章小结 21

第四章 南京市公共自行车通勤出行特征分析 22

4.1 通勤量特征分析 22

4.1.1 通勤出行总量特征分析 22

4.1.2 用户通勤出行量分析 22

4.2 通勤出行时长特征分析 25

4.2.1 不同通勤模式的通勤出行时长比较分析 25

4.2.2 用户通勤出行时长特征分析 25

4.3 通勤出行距离特征分析 27

剩余内容已隐藏，请支付后下载全文，论文总字数：49695字