Degree
Doctor of Philosophy (PhD)
Department
Bert S. Department of Construction
Document Type
Dissertation
Abstract
The construction industry faces a critical shortage of skilled heavy equipment operators while simultaneously requiring enhanced operational efficiency and reduced environmental impact, creating an urgent need for innovative approaches to operator training and performance optimization. This dissertation presents a comprehensive multi-domain approach to excavator performance optimization, with a primary focus on improving operator skill development and equipment training effectiveness through the integration of fill factor optimization, objective skill assessment, and intelligent feedback systems. Using a CAT excavator simulator as the experimental platform, this research investigates the relationships between optimal operating parameters and human expertise to establish a data-driven framework for operator training and performance enhancement. The first phase examines the impact of varying fill factors on fuel efficiency through controlled experiments comparing bucket normal and bucket return excavation approaches across multiple fill factors, with ANOVA analysis revealing that 70% fill factor represents a critical efficiency threshold for establishing evidence-based training guidelines. The second phase employs Pareto frontier analysis to examine trade-offs between productivity and greenhouse gas emissions across varying fill factors in excavation tasks, identifying optimal fill factor ranges of 80%-106% depending on operational priorities and providing benchmarks for training programs that balance productivity with environmental responsibility. The third phase introduces an objective skill assessment framework that analyzes joystick control signals to evaluate operator expertise without subjective human evaluation, using signal processing techniques and machine learning algorithms to achieve 99.13% accuracy in distinguishing between novice, intermediate, and expert operators, thereby eliminating training assessment bias while providing quantitative metrics for certification programs and personalized skill development pathways. The fourth phase integrates these findings into an intelligent feedback system that provides real-time guidance to operators during training exercises, combining optimal performance parameters with individual skill assessment data to deliver personalized recommendations that accelerate skill acquisition while promoting efficient and environmentally conscious operating practices. The integration of these components establishes a unified framework for excavator operator training that addresses the industry's workforce development challenges through objective assessment, personalized feedback, and evidence-based performance optimization, demonstrating that expert operator behavior patterns naturally align with optimal efficiency parameters and enabling the development of training systems that simultaneously improve human skills and equipment performance. This research contributes significantly to construction workforce development by providing practical solutions for training institutions, equipment manufacturers, and construction companies seeking to address operator shortages while improving operational efficiency and environmental performance, with the objective assessment and feedback systems offering scalable approaches to skill development that can transform how heavy equipment operators are trained and certified in an increasingly technology-driven industry.
Date
1-14-2026
Recommended Citation
Willoughby, Sueed A., "Data-Driven Construction Equipment Operation: Efficiency, Emissions, and Skill Classification" (2026). LSU Doctoral Dissertations. 6991.
https://repository.lsu.edu/gradschool_dissertations/6991
Committee Chair
Dr. Chao Wang