Data pertaining to Chapter 3 "Investigation on Car-Following Heterogeneity and Its Impacts on Traffic Flow Performance"

This dataset supports the paper <em>“Investigation on Car-Following Heterogeneity and Its Impacts on Traffic Flow Performance”</em> (Chapter 3 of the PhD dissertation). The study focuses on behavioural modelling and simulation, aiming to investigate car-following heterogeneity and assess its effects on traffic safety and sustainability. The framework incorporates rigorous driving style classification using a semi-supervised learning technique and a micro-simulation process that includes 66 fine-grained traffic scenarios exhibiting varying degrees of heterogeneity. Based on two distinct real-world datasets, the impacts of driving heterogeneity are effectively elucidated from the mechanism of underlying characteristics of driving behaviour and traffic flow dynamics. The data is organised into three folders, corresponding to model parameter calibration, behavioural classification, and traffic flow simulation components of the research. The data was generated and processed using MATLAB and includes files in <code>.xlsx</code>, <code>.csv</code>, <code>.mat</code>, <code>.m</code>, <code>.txt</code>, and <code>.pdf</code> formats. A <code>ch3_Readme.txt</code> file is provided to guide users through the dataset structure and usage.

本数据集为论文《跟驰行为异质性及其对交通流运行绩效的影响研究》（博士学位论文第3章）提供数据支撑。本研究聚焦行为建模与仿真方向，旨在探究跟驰行为异质性，并评估其对交通安全与交通可持续性的影响。该研究框架采用半监督学习技术实现严谨的驾驶风格分类，并配套包含66种不同异质性程度的精细化交通场景的微观仿真流程。本研究基于两组独立的真实世界数据集，从驾驶行为固有特征与交通流动力学机制层面，清晰阐明了驾驶异质性带来的影响。数据集按三个文件夹进行整理，分别对应本研究中的模型参数校准、行为分类与交通流仿真三大研究模块。本数据集通过MATLAB生成并处理，包含.xlsx、.csv、.mat、.m、.txt及.pdf多种格式的文件。附带ch3_Readme.txt文件，用于指导用户了解数据集的结构与使用方法。