Scholarlink

In pursuit of enhancing urban sustainability by mitigating traffic congestion, traditional efforts typically focus on optimizing transportation infrastructure. This paper introduces an innovative predictive framework for dynamic traffic assignment aimed at individual-level mobility optimization. By leveraging large-scale mobile phone data and road networks, travel demands are modeled and inter-zone traffic is forecasted using advanced deep learning techniques. These forecasts are integrated into a dynamic traffic assignment model, resulting in an iterative optimization framework that recommends socially optimal routes for individual travelers. Comprehensive experiments conducted in four Chinese cities demonstrate a significant reduction in road density and congestion during peak hours. To address the limitations of estimating road demand using mobile phone data, the framework incorporates base station deviation correction and sample expansion. This research makes a substantial contribution to urban sustainability by effectively alleviating congestion across the entire urban network.

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

Urban Traffic Management: A Predictive Approach Using Mobile Phone Data

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Xiaolong Niu,
  • Huachao Gao,
  • Wentao Hu,
  • Long Yan,
  • Yawei Ren,
  • Yang Li,
  • Dazhong Li,
  • Hao Ge,
  • Chunxiao Li,
  • Yanyan Xu,
  • Yulun Song

摘要

In pursuit of enhancing urban sustainability by mitigating traffic congestion, traditional efforts typically focus on optimizing transportation infrastructure. This paper introduces an innovative predictive framework for dynamic traffic assignment aimed at individual-level mobility optimization. By leveraging large-scale mobile phone data and road networks, travel demands are modeled and inter-zone traffic is forecasted using advanced deep learning techniques. These forecasts are integrated into a dynamic traffic assignment model, resulting in an iterative optimization framework that recommends socially optimal routes for individual travelers. Comprehensive experiments conducted in four Chinese cities demonstrate a significant reduction in road density and congestion during peak hours. To address the limitations of estimating road demand using mobile phone data, the framework incorporates base station deviation correction and sample expansion. This research makes a substantial contribution to urban sustainability by effectively alleviating congestion across the entire urban network.