Scholarlink

<p>With the development of intelligent technology, the analysis of travel time plays a key role in transportation decision-making. In response to the problems of low efficiency in trajectory data processing and low accuracy in travel time prediction, this study gridded the trajectory data of the Hangzhou traffic management department and optimized the data gridding process by combining timestamps. Finally, a fast travel time model was constructed using the Burr-III distribution model and comprehensively analyzed using multiple reliability indicators. The results indicated that the average travel time was relatively high for the roadway segments studied at 10 o’clock. This can prompt managers to take corresponding traffic diversion measures during peak hours to improve traffic efficiency. The average travel time of the road segment <i>Link 1</i> varied more steadily from 6 to 22 o’clock, indicating a higher level of reliability for <i>Link 1</i>. Passengers traveling on this section can better predict their travel time and make reasonable travel plans. The coefficient of variation for <i>Link 3</i> increased and then decreased from 6 to 12 o’clock, with a lower level of reliability. The coefficient of variation reached a maximum of 0.725 at 10 points. The coefficient of variation of <i>Link 5</i> at 18 points was the largest, reaching 0.92. Passengers need to reserve more time when traveling on these road sections to cope with possible delays. Whereas the Burr-III distribution curve fits well and the lognormal distribution performs poorly, the Burr-III distribution better describes the distribution of fast travel time of the sample data. The number of samples gradually increased with fast travel time, and the Burr-III distribution was more able to match the actual sample data. The distribution model had a better fit for describing the fast travel time probability distribution function. During the study period, the Kolmogorov-Smirnov test results of the Burr-III distribution model were 0 for different routes and road segments. This model can fit actual data well. The study improves the efficiency and accuracy of traffic data analysis through rasterized trajectory data processing and the application of the Burr-III distribution model, thus providing stronger support for real-time traffic management and travel decision-making.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Fast Travel Time Modeling Based on Raster Trajectory Data

  • 期刊论文

  • 发表日期：

    2025年8月4日

查看全文
  • Shasha Song,
  • Cong Li

摘要

With the development of intelligent technology, the analysis of travel time plays a key role in transportation decision-making. In response to the problems of low efficiency in trajectory data processing and low accuracy in travel time prediction, this study gridded the trajectory data of the Hangzhou traffic management department and optimized the data gridding process by combining timestamps. Finally, a fast travel time model was constructed using the Burr-III distribution model and comprehensively analyzed using multiple reliability indicators. The results indicated that the average travel time was relatively high for the roadway segments studied at 10 o’clock. This can prompt managers to take corresponding traffic diversion measures during peak hours to improve traffic efficiency. The average travel time of the road segment Link 1 varied more steadily from 6 to 22 o’clock, indicating a higher level of reliability for Link 1. Passengers traveling on this section can better predict their travel time and make reasonable travel plans. The coefficient of variation for Link 3 increased and then decreased from 6 to 12 o’clock, with a lower level of reliability. The coefficient of variation reached a maximum of 0.725 at 10 points. The coefficient of variation of Link 5 at 18 points was the largest, reaching 0.92. Passengers need to reserve more time when traveling on these road sections to cope with possible delays. Whereas the Burr-III distribution curve fits well and the lognormal distribution performs poorly, the Burr-III distribution better describes the distribution of fast travel time of the sample data. The number of samples gradually increased with fast travel time, and the Burr-III distribution was more able to match the actual sample data. The distribution model had a better fit for describing the fast travel time probability distribution function. During the study period, the Kolmogorov-Smirnov test results of the Burr-III distribution model were 0 for different routes and road segments. This model can fit actual data well. The study improves the efficiency and accuracy of traffic data analysis through rasterized trajectory data processing and the application of the Burr-III distribution model, thus providing stronger support for real-time traffic management and travel decision-making.