Rochester Institute of Technology
Leveraging Machine Learning for Traffic Congestion Management in Smart Cities
Abstract
dc:description.abstractTraffic congestion continues to be a major urban issue, leading to traffic delays, higher fuel costs, and air pollution problems. Traffic management systems currently function in reactive mode because their algorithms only operate following congestion development rather than preventing it. Smart cities need predictive systems based on data analytics and machine learning to actively control urban traffic movements because traffic continues to rise as a result of urbanization and population growth. The proposed research designs a machine learning–driven traffic congestion prediction system that uses genuine data obtained from Aarhus, Denmark, and METR-LA, Los Angeles. The study will analyze fundamental traffic data characteristics, including vehicle speed and volume, along with daytime periods and road classification types, and evaluate weather factors and urban scheduled activities on congestion behaviour—the main objective centers on uniting diverse data sources to improve operational accuracy of predictive models. The proposed research will deliver traffic congestion predictions based on critical related parameters. The study will identify the parameters that need to be considered for accurate predictions and the significance of the parameters on the predictions. The proposed research will show merging traffic with weather conditions and public transportation to improve the accuracy of the predictions. The prediction model will be measured to demonstrate effectiveness for different urban environments. The study evaluates model effectiveness, generalizability, and application potential by implementing classical machine learning algorithms such as linear regression, Decision Trees, and Random Forests. The framework will lead to an operational system that acts as a foundation for data-based decision support in smart city traffic management and shows scalability and adaptability capabilities.
Degree
thesis:*- Name thesis:degree_name
- Professional Studies (MS)
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Graduate Programs & Research
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Alhasai, Omar
- Contributors dc:contributor
-
- Khalil Al Hussaeni
Subjects
dc:subject × 4Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.rit.edu/theses/12538