Estimation of CO2 Emissions in Transportation Systems Using Artificial Neural Networks, Machine Learning, and Deep Learning - A Comprehensive Approach

Aug 6, 2026

Upload by Ivy

This study focuses on estimating transportation system-related emissions in CO2 eq., accounting for socioeconomic and energy- and transportation-specific input variables. The proposed approach integrates artificial neural networks, machine learning, and deep learning models, demonstrating their effectiveness for forecasting transport emissions. Turkey serves as the case study, with model performance assessed using Turkey-based data and future projections generated via scenario analysis aligned with Turkey’s climate change mitigation strategies. Results also include a transport-type breakdown across road, air, marine, and rail.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages21

Reading Time53 min

AccessFree to download

Introduction

Climate change and greenhouse gas indicators

Transport emissions and decarbonization challenges

Climate risk adaptation and policy alignment (Paris Agreement, NDCs)

Emissions forecasting methods and prior research

Article Estimation of CO2 Emissions in Transportation Systems Using Artificial Neural Networks, Machine Learning, and Deep Learning: A Comprehensive Approach Seval Ene Yalçın Academic Editor: Ewelina Sendek-Matysiak Received: 27 January 2025 Revised: 25 February 2025 Accepted: 10 March 2025 Published: 11 March 2025 Citation: Ene Yalçın, S. Estimation of CO2 Emissions in Transportation Systems Using Artificial Neural Networks, Machine Learning, and Deep Learning: A Comprehensive Approach. Systems 2025, 13, 194 . [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)systems13030194 Copyright: © 2025 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://creativecommons.org/](https://creativecommons.org/)[ ](https://creativecommons.org/)[licenses/by/4.0/](licenses/by/4.0/)) . Department of Industrial Engineering, Bursa Uluda ˘g University, Görükle Campus, 16059 Bursa, Türkiye; [sevalene@uludag.edu.tr](sevalene@uludag.edu.tr) Abstract: This study focuses on estimating transportation system-related emissions in CO2 eq., considering several socioeconomic and energy-and transportation-related input variables. The proposed approach incorporates artificial neural networks, machine learning, and deep learning algorithms. The case of Turkey was considered as an example. Model performance was evaluated using a dataset of Turkey, and future projections were made based on scenario analysis compatible with Turkey’s climate change mitigation strategies. This study also adopted a transportation type-based analysis, exploring the role of Turkey’s road, air, marine, and rail transportation systems. The findings of this study indicate that the aforementioned models can be effectively implemented to predict transport emissions, concluding that they have valuable and practical applications in this field. Keywords: artificial neural networks; deep learning; machine learning; CO 2 emissions; transport systems; forecasting 1. Introduction The increase in global consciousness regarding environmental issues and pollution has positioned models of environmental indicators and future projections as pivotal strategic instruments. Climate change is one of the most significant environmental issues at present. Greenhouse gas (GHG) emissions are key indicators of climate change. GHGs comprise carbon dioxide (CO 2), methane (CH 4), nitrous oxide (N 2O), and fluorinated gases. Aggregate emissions can be articulated in terms of CO 2 equivalents [1] . Transport is acknowledged as a key enabler of global trade and development. As with other economic sectors, the transport sector encounters considerable challenges in the context of climate change, specifically the need to reduce its carbon emissions [2], as it has been identified as one of the three top global contributors to CO 2 emissions. In 2022, total transport emissions exhibited an increase of 2.1%, which was predominantly driven by growth in advanced economies. It is worth noting that this increase would have been more pronounced in the absence of the accelerated adoption of low-carbon vehicles [3] . On a global scale, although there has been some progress in recent years towards the decarbonization of road transport, particularly in urban transport, the decarbonization process is still in its initial stages for road freight, shipping, and aviation [2] . Effectively adapting to climate risk in the transport sector can be facilitated by integrating risk assessment and adaptation strategy planning into national adaptation plans and processes. This is crucial for the implementation of international agreements, such as the Paris Agreement [4] . In accordance with the Paris Agreement, parties are required to submit Nationally Determined Contributions (NDCs) to the United Nations Framework Convention on Climate Change (

FAQs

What does the study aim to estimate in transportation systems?

The study estimates transportation system-related emissions in CO2 equivalents using socioeconomic and energy/transport input variables.

Which modeling approaches are incorporated in the proposed method?

The approach combines artificial neural networks with machine learning and deep learning algorithms.

How is the case study for evaluating the models determined?

Turkey is used as the example, with model performance evaluated using a Turkey dataset and future projections produced through scenario analysis.

Aug 6, 2026

Upload by Ivy