Lu, Ruicheng (tongji university), Zhang, Hua (Tongji University), Zhang, Zijing (Tongji University)

Revealing Urban Functionality through Vehicle Dwell Patterns: A Data-Driven Approach

Scheduled for presentation during the Regular Session "S35a-Optimization, Control, and Learning for Efficient and Resilient ITS" (FR-LM-T35), Friday, November 21, 2025, 10:50−11:10, Surfers Paradise 2

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on October 18, 2025

Abstract

Vehicle dwell patterns offer valuable insights into both travel behavior and urban spatial functions. However, accurately capturing these patterns remains challenging due to the complex spatiotemporal behaviors of vehicles and the diverse characteristics of urban areas. This study introduces an innovative approach by utilizing License Plate Recognition (LPR) data to uncover vehicle dwell patterns and their correlation with urban functional attributes. By analyzing two weeks of LPR data from over 1,400 detectors in Shanghai, we propose a multi-scale adaptive slope-change detection algorithm to identify temporal thresholds in vehicle dwell times. Combined with satellite imagery and Point-of-Interest (POI) data, we categorize vehicle behaviors into four main types: non-stopping, short-duration stop, medium-duration stay, and long-duration parking. Using K-means++ clustering, four distinct dwell patterns are identified: rapid throughput, mixed-function transitional, commercial-service oriented, and long-term parking zones. These findings offer practical insights for improving traffic signal control, parking management, and urban planning by considering the specific functions of different urban zones.