Date of Award

3-25-2024

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

School of Information Technology: Information Systems

First Advisor

Dr. Rezgui Abdelmounaam

Abstract

This thesis explores the use of machine learning techniques for road infrastructure maintenance. We propose an innovative machine learning-based approach to improve the efficiency and effectiveness of road maintenance strategies. The focal point of this investigation is the development and implementation of a machine learning framework to enhance road quality monitoring. We use Long Short-Term Memory (LSTM) networks to accurately predict future road conditions and identify potential areas requiring maintenance before significant deterioration occurs. This predictive approach is designed to enable a shift from reactive to proactive road maintenance, optimizing the use of limited resources and improving overall road safety. The methodology of the research is structured in three phases: the creation of a prototype system for road condition data collection, the application of LSTM networks for predictive analysis, and the utilization of optimization techniques to guide effective maintenance decisions. By focusing on predictive accuracy and the strategic allocation of maintenance efforts, the study seeks to extend the lifespan of road infrastructure, reduce maintenance costs, and enhance the driving experience. This thesis is a contribution to the field of road infrastructure maintenance by introducing a predictive maintenance model that leverages advanced machine learning techniques. It aims to transform the traditional maintenance approach, providing a scalable and efficient solution to road infrastructure management challenges, with the potential to significantly influence policy and practice in infrastructure maintenance.KEYWORDS: Machine learning; Infrastructure maintenance; Proactive maintenance

Comments

Imported from Rachala_ilstu_0092N_12624.pdf

DOI

https://doi.org/10.30707/ETD2024.20240618063950705825.999934

Page Count

73

Download

Share

COinS