Date of Award

12-2025

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Civil Engineering

Committee Chair/Advisor

Tuyen Le

Committee Member

Kalyan R. Piratla

Committee Member

Kapil Chalil Madathil

Committee Member

Chao Fan

Abstract

This research develops data-driven and AI-enhanced approaches to improve worker safety in highway construction environments. The study integrates advanced analytics and auditory signal enhancement to identify high-risk worker behaviors and improve real-time safety warnings. The work consists of three core studies. The first study applies Sequential Pattern Mining and Social Network Analysis to more than 1,000 construction accident reports to identify high-risk worker actions and their hidden sequential relationships with accident types and consequences. These insights reveal sector-specific risk patterns, supporting the creation of targeted, data-informed safety interventions. The second study introduces an AI-based auditory enhancement model built on a Conformer-based Metric Generative Adversarial Network (CMGAN). Designed for noisy construction environments, the model enhances the clarity of safety-critical signals, including backup alarms and intrusion alerts. By combining local acoustic feature extraction, long-range temporal modeling, and adversarial training with perceptual loss functions, the model suppresses background noise while preserving essential warning cues, even under extreme noise conditions. The third study develops a real-time auditory enhancement prototype using the lightweight DeepFilterNet algorithm. Implemented on a low-cost Raspberry Pi 5 and designed for integration into portable hearing protection devices, the system delivers low-latency enhancement of safety-critical signals. Field tests and listener surveys demonstrate improved audibility and preserved signal characteristics, confirming the feasibility of real-time deployment in active highway construction sites. Overall, this research enhances the understanding of high-risk worker behaviors and demonstrates the practical application of AI-driven auditory safety technologies with the potential to improve overall safety practices in the highway construction industry.

Author ORCID Identifier

https://orcid.org/0000-0002-1846-0764

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.