Abstract

City traffic jams have become a major challenge for contemporary cities, causing delays, increased fuel consumption, and other environmental impacts. Conventional traffic signal controlling systems usually depend on fixed or preset signal plans that are incompetent to adjust themselves with changing traffic conditions. Traffic signal timings are optimized across the intersection health of the system by processing real-time data analytics. Through real-time data analytics, traffic flow efficiency is improved while waiting time at signalized intersections is reduced. Using traffic real-time data like vehicle counts, vehicle types, time of day and day of the week to see howtraffic behaves under different situations. Statistical and analytical methods such as Chi- Square test and exploratory data analysis will be used to find out correlation between various traffic variables and congestion. This paper recommends adaptive signal timing strategies based on real-time traffic demand profiles gained from this investigation. The findings indicate a clear variation over time in traffic characteristics, supporting the limitations of fixed signal timings. The predictive models have undergone performance evaluation. It has been found that the Neural Network has the highest accuracy. The correlation coefficient is 0.98. The mean relative error is 0.0416. The value of R2 is 0.97. It outperformed the XGBoost, Random Forest, LSVM, and Linear Regression models. The data-driven approach for traffic signal optimizationwas predictive in nature and supports the effectiveness of adaptive real-time traffic control strategy.

Publication Date

2-18-2026

Document Type

Thesis

Student Type

Graduate

Degree Name

Professional Studies (MS)

Department, Program, or Center

Graduate Programs & Research

Advisor

Hammou Messatfa

Campus

RIT Dubai

Download

Share

COinS