Travel Behavior Characteristics Analysis Technology Based on Mobile Phone Location Data: Methodology and Empirical Research

Preface
List of Figures
Contents
List of Tables
Foreword
1 Introduction
	1.1 General
		1.1.1 Drawbacks of Individual Travel Survey Methods
		1.1.2 Advantages of Mobile Phone Sensor Survey Methods
		1.1.3 Traffic Demand Analysis Model Development Challenges
		1.1.4 New Opportunities in the Era of Traffic Big Data
	1.2 Target and Values of Mobile Phone Data Based Travel Survey Method
	1.3 From Mobile Phone Location Data to Travel Information
		1.3.1 Mobile Phone Location Data Collection and Analysis
		1.3.2 Refined Travel Information Extraction and Collection
		1.3.3 ‘Man-Vehicle-Communication’ Simulation Platform Construction and Simulation
		1.3.4 Empirical Study and Performance Evaluation
		1.3.5 Challenges Faced by Travel Information Detection and Analysis
	1.4 Summary
2 Literature Review
	2.1 Types and Characteristics of Mobile Data Based Travel Survey Method
		2.1.1 Mobile Phone Sensor Data Based Travel Survey Method
		2.1.2 Mobile Phone Signaling Data Based Travel Survey Method
		2.1.3 Mobile Phone Social Network Data Based Travel Survey Method
	2.2 Overview and Summary of Existing Researches and Applications
	2.3 Individual Travel Behavior Analysis Based on Mobile Phone Signaling Data
		2.3.1 Dynamic Monitoring of Residents’ Travel Activities
		2.3.2 Regional and Cross-Section Passenger Flow Analysis
	2.4 Individual Travel Behavior Analysis Based on Mobile Phone Sensor Data
		2.4.1 Trip Chain Information Extraction
		2.4.2 Resident Travel Survey Application
	2.5 Activity Hotspots Analysis Based on Wi-Fi Data
	2.6 Individual Travel Behavior Analysis Based on Mobile Phone Social Network Data
		2.6.1 Resident Travel Characteristics Detection
		2.6.2 Trip OD Estimation
		2.6.3 Characteristics of Job and Residence Distribution
	2.7 Research Summary and Trend
	References
3 Methodology for Mobile Phone Location Data Mining
	3.1 Technology Structure for Individual Travel Chain Information Extraction
	3.2 Trip End Recognition Based on Spatial Clustering Algorithm
	3.3 Mode Transfer Point Recognition Based on Wavelet Analysis Algorithm
	3.4 Travel Mode Recognition Based on Machine Learning Algorithm
		3.4.1 Neural Network Algorithm
		3.4.2 Support Vector Machine Algorithm
		3.4.3 Bayesian Network Algorithm
		3.4.4 Random Forest Algorithm
	3.5 Trip Chain Information Optimization Based on GIS Map Matching
	3.6 Summary
	References
4 Mobile Phone Sensor Data Collection and Analysis
	4.1 Data Collection App Development
		4.1.1 Function Description
		4.1.2 Operation Interface
	4.2 Database Construction and Management
	4.3 Privacy and Data Security
	4.4 Characteristics Analysis of Mobile Phone Sensor Data
		4.4.1 GPS Data Accuracy and Quality
		4.4.2 Spatial–temporal Travel Characteristics
		4.4.3 Travel Trajectory Point Density
		4.4.4 Travel Speed Characteristics
		4.4.5 Travel Acceleration Characteristics
	4.5 Summary
5 ‘Pedestrian-Traffic Flow-Communication’ Integrated Simulation Platform Construction
	5.1 Framework of the Simulation Platform
	5.2 Traffic Environment and Individual Travel Simulation
		5.2.1 Traffic Environment Design
		5.2.2 Individual Travel Module Construction and Simulation
	5.3 Wireless Communication Simulation
		5.3.1 Wireless Communication Events Description and Simulation
		5.3.2 Mobile Communication Signal Propagation Simulation
		5.3.3 A Case Study of Wireless Communication Simulation
	5.4 Mobile Phone Sensor Data Simulation
		5.4.1 Data Disturbance Loading Method and Simulation
		5.4.2 A Case Study of Mobile Phone Sensor Data Simulation
	5.5 Summary
	References
6 Empirical Study on Trip Information Extraction Based on Mobile Phone Sensor Data
	6.1 Experiment Design and Data Collection
		6.1.1 Travel Plan for Different Travel Purposes
		6.1.2 Travel Plan for Multiple Modes
		6.1.3 Travel Plan for Different Traffic Conditions
		6.1.4 Travel Log Collection
	6.2 Empirical Study of Trip End Recognition Based on Spatial Clustering Algorithm
		6.2.1 Model Parameter Configuration
		6.2.2 A Case Study of Trip End Recognition and Travel Trajectory Cutting
		6.2.3 Results and Error Analysis
	6.3 Empirical Study of Mode Transfer Point Recognition Based on Wavelet Transform Modulus Maximum Algorithm
		6.3.1 Model Parameter Configuration
		6.3.2 A Case Study of Mode Transfer Point Recognition
		6.3.3 Results and Error Analysis
	6.4 Empirical Study of Travel Mode Recognition Based on Neural Network Algorithm
		6.4.1 Model Parameter Configuration
		6.4.2 A Case Study of Traffic Mode Recognition
		6.4.3 Results and Error Analysis
	6.5 Empirical Study of Travel Chain Recognition Optimization Based on GIS Map Matching
		6.5.1 Model Parameter Configuration
		6.5.2 A Case Study of Travel Mode Recognition Optimization
		6.5.3 Results and Error Analysis
	6.6 Summary
7 Influence Parameters and Sensitivity Analysis
	7.1 Influencing Factors and Mechanism
	7.2 Data Characteristics Under Different Experiment Conditions
		7.2.1 Data Collection
		7.2.2 Data Analysis
	7.3 Sensitivity Analysis of Travel Mode Recognition
		7.3.1 Influence of Algorithms
		7.3.2 Influence of Data Sampling Frequency
		7.3.3 Influence of Traffic Condition
	7.4 Sensitivity Analysis of Mode Transfer Time Recognition
		7.4.1 Influence of Algorithm
		7.4.2 Influence of Data Sampling Frequency
		7.4.3 Influence of Traffic Condition
	7.5 Sensitivity Analysis of Trip Chain Information Recognition Based on Simulation Data
		7.5.1 Sensitivity Analysis of Travel Mode
		7.5.2 Sensitivity Analysis of Mode Transfer Time
	7.6 Summary
8 Thinking About Application of Refined Travel Data in Traffic Planning
	8.1 Optimizing the Traditional Four-Step Method
	8.2 Optimizing the Layout of Bus Stations and Network
	8.3 Constructing Activity Based Traffic Demand Model
	8.4 Other Applications
9 Outlook
	9.1 Technical Efficiency and Universal Upgrading
	9.2 Multiple Heterogeneous Data Integrating
	9.3 Traffic Planning Theories and Models Upgrading