Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Estimating On-road Transportation Carbon Emissions from Open Data of Road Network and Origin-destination Flow Data

Jinwei Zeng, Yu Liu, Jingtao Ding, Jian Yuan, Yong Li

TL;DR

Estimating on-road transportation carbon emissions is challenging due to reliance on hard-to-collect vehicle miles data. This work introduces HENCE, a hierarchical heterogeneous graph learning framework that uses open origin-destination flow data and road-network data to model multi-scale interactions between travel demand and network connectivity. Across two large real-world datasets, HENCE achieves $R^2$ exceeding $0.75$ and outperforms baselines by about $9.6\%$ on average, with ablation showing the necessity of both the hierarchical structure and the heterogeneous edges. The approach demonstrates the potential of AI-powered open-data methods for carbon-emission monitoring and policy-support in transportation sustainability, including transferable performance across spatial and temporal domains.

Abstract

Accounting for over 20% of the total carbon emissions, the precise estimation of on-road transportation carbon emissions is crucial for carbon emission monitoring and efficient mitigation policy formulation. However, existing estimation methods typically depend on hard-to-collect individual statistics of vehicle miles traveled to calculate emissions, thereby suffering from high data collection difficulty. To relieve this issue by utilizing the strong pattern recognition of artificial intelligence, we incorporate two sources of open data representative of the transportation demand and capacity factors, the origin-destination (OD) flow data and the road network data, to build a hierarchical heterogeneous graph learning method for on-road carbon emission estimation (HENCE). Specifically, a hierarchical graph consisting of the road network level, community level, and region level is constructed to model the multi-scale road network-based connectivity and travel connection between spatial areas. Heterogeneous graphs consisting of OD links and spatial links are further built at both the community level and region level to capture the intrinsic interactions between travel demand and road network accessibility. Extensive experiments on two large-scale real-world datasets demonstrate HENCE's effectiveness and superiority with R-squared exceeding 0.75 and outperforming baselines by 9.60% on average, validating its success in pioneering the use of artificial intelligence to empower carbon emission management and sustainability development. The implementation codes are available at this link: https://github.com/tsinghua-fib-lab/HENCE.

Estimating On-road Transportation Carbon Emissions from Open Data of Road Network and Origin-destination Flow Data

TL;DR

Estimating on-road transportation carbon emissions is challenging due to reliance on hard-to-collect vehicle miles data. This work introduces HENCE, a hierarchical heterogeneous graph learning framework that uses open origin-destination flow data and road-network data to model multi-scale interactions between travel demand and network connectivity. Across two large real-world datasets, HENCE achieves exceeding and outperforms baselines by about on average, with ablation showing the necessity of both the hierarchical structure and the heterogeneous edges. The approach demonstrates the potential of AI-powered open-data methods for carbon-emission monitoring and policy-support in transportation sustainability, including transferable performance across spatial and temporal domains.

Abstract

Accounting for over 20% of the total carbon emissions, the precise estimation of on-road transportation carbon emissions is crucial for carbon emission monitoring and efficient mitigation policy formulation. However, existing estimation methods typically depend on hard-to-collect individual statistics of vehicle miles traveled to calculate emissions, thereby suffering from high data collection difficulty. To relieve this issue by utilizing the strong pattern recognition of artificial intelligence, we incorporate two sources of open data representative of the transportation demand and capacity factors, the origin-destination (OD) flow data and the road network data, to build a hierarchical heterogeneous graph learning method for on-road carbon emission estimation (HENCE). Specifically, a hierarchical graph consisting of the road network level, community level, and region level is constructed to model the multi-scale road network-based connectivity and travel connection between spatial areas. Heterogeneous graphs consisting of OD links and spatial links are further built at both the community level and region level to capture the intrinsic interactions between travel demand and road network accessibility. Extensive experiments on two large-scale real-world datasets demonstrate HENCE's effectiveness and superiority with R-squared exceeding 0.75 and outperforming baselines by 9.60% on average, validating its success in pioneering the use of artificial intelligence to empower carbon emission management and sustainability development. The implementation codes are available at this link: https://github.com/tsinghua-fib-lab/HENCE.
Paper Structure (20 sections, 11 equations, 3 figures, 4 tables)

This paper contains 20 sections, 11 equations, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Method
  4. Model Framework
  5. Heterogeneous Graph Learning Module for Road Network and OD Interaction Modeling
  6. Hierarchical Graph Learning Module for Multi-scale Information Propagation
  7. Multi-scale Information Fusion and Prediction
  8. Experiment
  9. Experimental Setups
  10. Dataset
  11. Baselines
  12. Metrics and Implementation
  13. Overall Performance
  14. Ablation Study
  15. Transferability Study
  16. ...and 5 more sections

Figures (3)

  • Figure 1: Visualization of road network length and per capita transportation demand for counties in the United States. Here the color corresponds to the county's on-road carbon emissions. Data sources are consistent with the experimental setting.
  • Figure 2: Our hierarchical heterogeneous graph learning method for on-road carbon emission estimation.
  • Figure 3: Message aggregation weights for OD links and spatial links of the region level and community level.