Date of Award
6-12-2025
Document Type
Thesis
Degree Name
Engineering, MSE
First Advisor
Zahid Hossain
Committee Members
Alan Meadors; Ashraf Elsayed
Call Number
ISBN 9798280760103
Abstract
Crash severity is a significant aspect of transport safety as its contributing factors assist engineers and designers in designing safer roads. The focus of this project is to predict crash severity, analyze crash parameters, and modeling of the 85th percentile speed (V85) selected rural roads (i.e., Interstate-555, East Johnson Avenue Highway, and Red Wolf Blvd) in Arkansas, using Artificial Neural Network (ANN) models. MATLAB® was used to predict the V85 and compare it with the actual V85 collected from field instrumentation. Besides the speed data, weather (e.g., rainfall), road geometry, light conditions, and traffic volume were used as input in training the ANN model. In addition, the project employs the Federal Highway Administration’s (FHWA) USLIMITS2 software for calculating the data-based speed limits. This study found that there is a strong correlation between higher speed and severity of crashes, which establishes the need for predictive tools to enhance traffic safety policy.
Rights Management
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Basel, Sagun, "Predictive Analysis of Crash Severity and Modeling Of 85th Percentile Speed for Rural Highways of Arkansas Using Artificial Intelligence (AI)" (2025). Student Theses and Dissertations. 1073.
https://arch.astate.edu/all-etd/1073
