Optimizing Traffic Signal Control Using Machine Learning and Environmental Data

Edidiong E. Akpan; Oluwatobi Akinlade; Oluwaseyi O. Alabi; Oluwasesan A. David; Oluseyi F. Afe; Sunday Adeola Ajagbe

doi:10.47852/bonviewAIA52024199

Authors

Edidiong E. Akpan School of Computing and Informatics, University of Louisiana at Lafayette, USA
Oluwatobi Akinlade Department of Computer Science, Birmingham City University, UK
Oluwaseyi O. Alabi Department of Mechanical Engineering, Lead City University, Nigeria https://orcid.org/0009-0005-0027-5930
Oluwasesan A. David Department of Computer Science, Nottingham Trent University, UK
Oluseyi F. Afe Department of Computer Science, Lead City University, Nigeria
Sunday Adeola Ajagbe Department of Computer Engineering, Abiola Ajimobi Technical University, Nigeria and Department of Computer Science, University of Zululand, South Africa https://orcid.org/0000-0002-7010-5540

DOI:

https://doi.org/10.47852/bonviewAIA52024199

Keywords:

machine learning, forecasting, environmental data, urbanization, traffic, air pollution

Abstract

Traffic signal control is a critical component of urban transportation management, and optimizing its performance can significantly reduce congestion, decrease travel times, and improve air quality. This study proposes a novel approach to optimizing traffic signal control using machine learning and environmental data. This work focuses on the interplay between smart city infrastructure and environmental data to provide a novel method for traffic pattern prediction. Mitigating traffic congestion is a pressing concern in urbanized societies and emerging smart cities. This study explores leveraging publicly available air pollution data as an environmental indicator to enhance urban mobility and predict traffic patterns. Taking into account factors including vehicle emissions, weather patterns, and topographical features, the study will look at possible connections between air pollution and traffic congestion. The goal of this project is to develop a prediction model that uses real-time air quality data for traffic forecasting by utilizing big data analytics and machine learning approaches. According to our research, the K-nearest neighbors (KNN) model performs better than any other regression model examined. According to experimental findings, the KNN model considerably lowers the error rate in traffic congestion prediction by over 30%.

Received: 29 August 2024 | Revised: 1 July 2025 | Accepted: 17 October 2025

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Author Contribution Statement

Edidiong E. Akpan: Conceptualization, Supervision, Project administration. Oluwatobi Akinlade: Data curation, Writing – original draft. Oluwaseyi O. Alabi: Methodology, Formal analysis, Resources, Data curation, Visualization. Oluwasesan A. David: Software, Validation. Oluwaseyi F. Afe: Software, Investigation. Sunday Adeola Ajagbe: Conceptualization, Resources, Writing – review & editing.