Document Type

Thesis

Date of Award

5-2023

School/College

College of Science, Engineering, and Technology (COSET)

Degree Name

MS in Transportation Planning & Management

Committee Chairperson

Mehdi Azimi

Committee Co-Chairperson

Yi Qi

Committee Member 1

Fengxiang Qiao, S

Committee Member 2

Subasish Das

Committee Member 3

Ismet Sahin

Keywords

• Artificial Intelligence • COVID-19 • Machine Learning • Operating Speed • Traffic Mobility

Abstract

Having large number of vehicles operating in the freeways of Houston daily, the mobility concern is high as some of the freeways in Houston are among the most congested freeways in United States. During the COVID-19 pandemic, the less congested freeways led to over speeding resulting in various crashes and even fatality. This resulted in changing of drivers; and ultimately the mobility patterns were changed during the study years of 2019, 2020 and 2021. To better understand how this mobility pattern changed over the three years, this research used Machine Learning algorithms to examine the mobility of freeways in Houston during that time. For this purpose, a model was developed using python coding which considered operating speed and other independent variables to understand the change of the traffic mobility. Several methods were used in the study to check the effectiveness of Artificial Intelligence modeling. To check how the mobility was impacted over the years, Violin Plots were also plotted to illustrate the change of operating speed from year 2019 to 2021. The results of this research demonstrated that there are eight factors that have significant effects on the vehicular mobility. Among them, annual average daily traffic is the most influencing in traffic mobility study whereas K-factor is the least effective among the selected variables. Relative countermeasures were recommended according to the influencing factors that were identified.

Share

COinS