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

Published on: February 07, 2024

Impact Score: 8.27

Arxiv code: Arxiv:2402.05153

Summary

• What is new: The research introduces a novel hierarchical heterogeneous graph learning method named HENCE for estimating on-road carbon emissions, leveraging AI and open data to overcome the challenges in collecting individual vehicle statistics.
• Why this is important: Existing methods for estimating on-road transportation carbon emissions rely heavily on hard-to-collect individual vehicle statistics, leading to high data collection difficulties.
• What the research proposes: The proposed solution, HENCE, utilizes open data sources, including origin-destination flow data and road network data, to construct a hierarchical heterogeneous graph that models the connectivity and interactions within a multi-scale road network. This approach leverages artificial intelligence for more accurate estimation of carbon emissions.
• Results: Experiments with two large-scale real-world datasets showed that HENCE achieved an R-squared exceeding 0.75 and improved estimation accuracy by 9.60% over existing baseline methods.

Technical Details

Technological frameworks used: Hierarchical heterogeneous graph learning

Models used: Not explicitly mentioned, but involves AI models for pattern recognition and estimation.

Data used: Origin-destination flow data and road network data

Potential Impact

Environmental monitoring agencies, policy-makers in urban planning and transportation, companies in the automotive and transportation sectors could benefit or need to adapt to these insights for better carbon emission management and sustainability planning.

Want to implement this idea in a business?

We have generated a startup concept here: EcoRoutes AI.