Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Estimating link level traffic emissions: enhancing MOVES with open-source data

Lijiao Wang, Muhammad Usama, Haris N. Koutsopoulos, Zhengbing He

TL;DR

This paper addresses the challenge of city-scale vehicle emissions estimation by replacing the MOVES operating-mode distribution step with a data-driven Modular Neural Network (MNN) trained on open-source data. The framework integrates OPEN data streams—OSM road networks, open GPS traces, regional traffic data, and satellite imagery feature vectors—and uses a ground-truth from public trajectories to learn a 23-class distribution over MOVES operating modes. The MNN demonstrates substantial improvements over the MOVES default method, achieving over 50% RMSE reduction across CO, NOx, CO2, and PM2.5, and higher $R^2$ and lower MAPE when estimating emissions for 1069 test links in 45 Boston-area towns. The approach offers a scalable, low-cost, and replicable pathway for large-scale emissions modeling using only open data, with potential extension to other cities and further enhancements from satellite-imagery-based representations of urban form.

Abstract

Open-source data offers a scalable and transparent foundation for estimating vehicle activity and emissions in urban regions. In this study, we propose a data-driven framework that integrates MOVES and open-source GPS trajectory data, OpenStreetMap (OSM) road networks, regional traffic datasets and satellite imagery-derived feature vectors to estimate the link level operating mode distribution and traffic emissions. A neural network model is trained to predict the distribution of MOVES-defined operating modes using only features derived from readily available data. The proposed methodology was applied using open-source data related to 45 municipalities in the Boston Metropolitan area. The "ground truth" operating mode distribution was established using OSM open-source GPS trajectories. Compared to the MOVES baseline, the proposed model reduces RMSE by over 50% for regional scale traffic emissions of key pollutants including CO, NOx, CO2, and PM2.5. This study demonstrates the feasibility of low-cost, replicable, and data-driven emissions estimation using fully open data sources.

Estimating link level traffic emissions: enhancing MOVES with open-source data

TL;DR

This paper addresses the challenge of city-scale vehicle emissions estimation by replacing the MOVES operating-mode distribution step with a data-driven Modular Neural Network (MNN) trained on open-source data. The framework integrates OPEN data streams—OSM road networks, open GPS traces, regional traffic data, and satellite imagery feature vectors—and uses a ground-truth from public trajectories to learn a 23-class distribution over MOVES operating modes. The MNN demonstrates substantial improvements over the MOVES default method, achieving over 50% RMSE reduction across CO, NOx, CO2, and PM2.5, and higher and lower MAPE when estimating emissions for 1069 test links in 45 Boston-area towns. The approach offers a scalable, low-cost, and replicable pathway for large-scale emissions modeling using only open data, with potential extension to other cities and further enhancements from satellite-imagery-based representations of urban form.

Abstract

Open-source data offers a scalable and transparent foundation for estimating vehicle activity and emissions in urban regions. In this study, we propose a data-driven framework that integrates MOVES and open-source GPS trajectory data, OpenStreetMap (OSM) road networks, regional traffic datasets and satellite imagery-derived feature vectors to estimate the link level operating mode distribution and traffic emissions. A neural network model is trained to predict the distribution of MOVES-defined operating modes using only features derived from readily available data. The proposed methodology was applied using open-source data related to 45 municipalities in the Boston Metropolitan area. The "ground truth" operating mode distribution was established using OSM open-source GPS trajectories. Compared to the MOVES baseline, the proposed model reduces RMSE by over 50% for regional scale traffic emissions of key pollutants including CO, NOx, CO2, and PM2.5. This study demonstrates the feasibility of low-cost, replicable, and data-driven emissions estimation using fully open data sources.

Paper Structure

This paper contains 18 sections, 2 equations, 15 figures, 1 table.

Table of Contents

  1. Abstract
  2. Introduction
  3. METHODOLOGY
  4. Model
  5. Data
  6. Infrastructure
  7. Traffic
  8. Satellite Imagery
  9. Training
  10. Trajectory Data
  11. Application
  12. Data preprocessing
  13. Training
  14. Evaluation
  15. Conclusion
  16. ...and 3 more sections

Figures (15)

  • Figure 1: Methodological framework
  • Figure 2: Modular neural network architecture with input nodes, hidden layers, and output layer with drop out ratio, adopted from wang2025estimating.
  • Figure 3: A framework that encodes satellite imagery into feature vector.
  • Figure 4: Workflow for preprocessing OSM GPS trajectories, including map matching using Valhalla and smoothing with local regression.
  • Figure 5: Raw GPS Positions and Smoothed Curve in One Second
  • ...and 10 more figures