Estimating Highway Rehabilitation Projects Using Machine Learning

Authors

• Samantha Ocaña, The University of Texas Rio Grande Valley
• Silvia Flores-Osuna, The University of Texas Rio Grande Valley
• Gasser G. Ali, The University of Texas Rio Grande Valley
• Constantine Tarawneh, The University of Texas Rio Grande ValleyFollow

Document Type

Conference Proceeding

Publication Date

8-27-2026

Abstract

Highways are a crucial part of the national infrastructure. As such, they are costly to construct and maintain; challenges such as deterioration, limited budgets, and cost uncertainty create a need for improved cost prediction of rehabilitation projects. This is especially true in the preliminary stage when all factors are unknown. Accordingly, this paper proposes a model to predict the cost of highway rehabilitation projects. A machine learning model is trained on large and publicly available datasets released by the Texas Department of Transportation (TxDOT), the National Oceanic and Atmospheric Administration (NOAA), and the Federal Reserve of Economic Data (FRED). These datasets include past projects, along with their maintenance dates, costs, annual average daily traffic, temperature data, and cost indices. Several models were tested, and XGBoost demonstrated the best performance, achieving an R² of 0.74. Further inspection of the inputs showed that the type of project and project length were the strongest predictors of project cost, with AADT and cost indices also contributing. Temperature variables had little to no influence. The findings show that an XGBoost model can provide reliable cost predictions, supporting informed decision-making and more efficient resource allocation in future highway rehabilitation projects.

Comments

© 2026 American Society of Civil Engineers. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.

Recommended Citation

Ocaña, Samantha, Silvia Flores-Osuna, Gasser G. Ali, and Constantine Tarawneh. "Estimating highway rehabilitation projects using machine learning." In Construction Research Congress 2026, pp. 147-156. 2026. https://doi.org/10.1061/9780784486962.015

First Page

147

Publication Title

Construction Research Congress 2026 Advanced Technologies Artificial Intelligence and Data Analytics in Construction Selected Papers from Construction Research Congress 2026

Streaming Media

DOI

10.1061/9780784486962.015