Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Exploring Urban Mobility Trends using Cellular Network Data

Oluwaleke Yusuf, Adil Rasheed, Frank Lindseth

TL;DR

The paper tackles the challenge of obtaining detailed, scalable mobility data in Trondheim by leveraging cellular network signals from Telia Crowd Insights to generate routing reports. It develops a preprocessing and graph-augmentation framework to convert anonymized telecom signals into analyzable mobility metrics, enabling geospatial and temporal analyses across multiple transportation modes, while accounting for external factors such as COVID-19, weather, and road attributes. Key contributions include a comprehensive dataset description, data-cleaning and feature-engineering pipelines, and route- and area-specific mobility insights that inform MoST and Miljøpakken planning, with validation against APC-based public transit data. The work demonstrates the potential of cellular-network-derived mobility data to support data-driven urban planning and digital twin initiatives, while highlighting privacy considerations and avenues for future extension to cycling, walking, and predictive spatiotemporal modelling.

Abstract

The growth of urban areas intensifies the need for sustainable, efficient transportation infrastructure and mobility systems, driving initiatives to enhance infrastructure and public transit while reducing traffic congestion and emissions. By utilizing real-world data, a data-driven approach can provide crucial insights for urban mobility planning and decision-making. This study explores the efficacy of leveraging telecoms data from cellular network signals for studying crowd movement patterns, focusing on Trondheim, Norway. It examines routing reports to understand the spatiotemporal dynamics of various transportation routes and modes. A data preprocessing and feature engineering framework was developed to process raw routing reports for historical analysis. This enabled the examination of geospatial trends and temporal patterns, including a comparative analysis of various transportation modes, along with public transit usage. Specific routes and areas were analyzed in-depth to compare their mobility patterns with the broader city context. The study highlights the potential of cellular network data as a resource for shaping urban transportation and mobility systems. By identifying deficiencies and potential improvements, city planners and stakeholders can foster more sustainable and effective transportation and mobility solutions.

Exploring Urban Mobility Trends using Cellular Network Data

TL;DR

The paper tackles the challenge of obtaining detailed, scalable mobility data in Trondheim by leveraging cellular network signals from Telia Crowd Insights to generate routing reports. It develops a preprocessing and graph-augmentation framework to convert anonymized telecom signals into analyzable mobility metrics, enabling geospatial and temporal analyses across multiple transportation modes, while accounting for external factors such as COVID-19, weather, and road attributes. Key contributions include a comprehensive dataset description, data-cleaning and feature-engineering pipelines, and route- and area-specific mobility insights that inform MoST and Miljøpakken planning, with validation against APC-based public transit data. The work demonstrates the potential of cellular-network-derived mobility data to support data-driven urban planning and digital twin initiatives, while highlighting privacy considerations and avenues for future extension to cycling, walking, and predictive spatiotemporal modelling.

Abstract

The growth of urban areas intensifies the need for sustainable, efficient transportation infrastructure and mobility systems, driving initiatives to enhance infrastructure and public transit while reducing traffic congestion and emissions. By utilizing real-world data, a data-driven approach can provide crucial insights for urban mobility planning and decision-making. This study explores the efficacy of leveraging telecoms data from cellular network signals for studying crowd movement patterns, focusing on Trondheim, Norway. It examines routing reports to understand the spatiotemporal dynamics of various transportation routes and modes. A data preprocessing and feature engineering framework was developed to process raw routing reports for historical analysis. This enabled the examination of geospatial trends and temporal patterns, including a comparative analysis of various transportation modes, along with public transit usage. Specific routes and areas were analyzed in-depth to compare their mobility patterns with the broader city context. The study highlights the potential of cellular network data as a resource for shaping urban transportation and mobility systems. By identifying deficiencies and potential improvements, city planners and stakeholders can foster more sustainable and effective transportation and mobility solutions.
Paper Structure (19 sections, 14 figures)

This paper contains 19 sections, 14 figures.

Table of Contents

  1. Introduction
  2. Cellular Network Data
  3. Methodology
  4. Advantages and Challenges
  5. Routing Reports for Trondheim Municipality
  6. Data Description
  7. Data Preprocessing, Feature Engineering, and Graph Augmentation
  8. Analyses and Discussion
  9. Geospatial Trends
  10. Temporal (Normalized) Patterns
  11. External Factors
  12. COVID-19 Pandemic
  13. Weather Conditions
  14. Speed Limits and Lane Counts
  15. Specific Routes and Areas
  16. ...and 4 more sections

Figures (14)

  • Figure 1: Satellite map of ways covered by the routing reports
  • Figure 2: Yearly breakdown of peopleFlow volumes from January 2019 to November 2023.
  • Figure 3: Detailed overview of peopleFlow volumes across municipalities
  • Figure 4: Detailed overview of peopleFlow volumes across transportation modes
  • Figure 5: Normalized hourly variation of peopleFlow volumes across transportation modes
  • ...and 9 more figures