2026 ASEE Annual Conference & Exposition

A Modular, Extensible AI-Based PPE Compliance Framework for Engineering Technology Labs

Presented at Engineering Technology Division (ETD) Technical Session 5

Ensuring consistent compliance with personal protective equipment (PPE) requirements in engineering technology laboratories is a persistent instructional and safety challenge. In many laboratory settings, PPE enforcement relies primarily on instructor observation, which can be difficult to maintain when multiple students are engaged in hands-on activities. As class sizes grow and laboratory tasks become more complex, opportunities for missed or delayed safety interventions increase. This paper presents a modular, extensible, AI-based PPE compliance framework designed to augment instructor oversight in instructional laboratories without replacing human instruction. The proposed system integrates an Intel RealSense camera with lightweight computer vision models to automatically assess PPE compliance at the individual student level. The current implementation focuses on two critical PPE elements—face masks and safety glasses—selected because they are commonly required across manufacturing, mechatronics, robotics, and electromechanical laboratories. A face detection module first identifies individuals in the scene. Dedicated deep learning models are then applied to each detected face region to determine mask usage and the presence of safety glasses. The system operates in real-time on standard laboratory computers and provides immediate visual feedback, allowing students to correct unsafe conditions as they occur. While the present work emphasizes mask and eye protection, the framework is intentionally designed to support incremental expansion to additional PPE categories, gesture recognition, and depth-based safety zone monitoring. Experimental observations demonstrate reliable real-time performance under typical laboratory conditions while also identifying practical limitations related to lighting, viewing angle, and PPE appearance. These findings suggest that AI-assisted PPE monitoring can serve as a practical instructional aid for reinforcing safe practices in engineering technology education.

Authors

Dr. Mahdi Yazdanpour Miami University [biography]

Dr. Mahdi Yazdanpour (Senior Member of IEEE, ABET PEV) is an Associate Professor at Miami University. He previously served as an Assistant Professor and the Program Coordinator of Mechatronics Engineering Technology, as well as the Coordinator of the Institute for Student Research and Creative Activity at Northern Kentucky University. During his academic journey, he has had the privilege of exploring various engineering disciplines. His educational path began with a B.S. in Computer Engineering, followed by an M.S. in Industrial Engineering, and a Ph.D. in Electrical Engineering. This diverse academic background has enriched his problem-solving abilities and equipped him with the capacity to approach engineering challenges from different perspectives. He has over 11 years of experience in teaching and research in academia, and over 10 years of professional experience in industry, where he held various technical, managerial, and leadership roles in manufacturing, power plants, and oil & gas companies. Dr. Yazdanpour has received numerous awards and recognitions, including the "2026 IEEE Region 2 Outstanding Engineering Educator Award", the "2023 College of Arts and Sciences Outstanding Junior Faculty Award" at Northern Kentucky University and the "College of Engineering, Architecture and Technology Outstanding Teaching Assistant Award" at Oklahoma State University.
Additionally, Dr. Yazdanpour is the recipient of several internal and external research grants, including the National Institutes of Health (NIH) Research Evaluation and Commercialization Hub (REACH) Product Development Grant through KYNETIC Cycle 5, NIH Mid-South REACH, and the Kentucky Commercialization Ventures (KCV) IMPACT Competition Award.
His research interests include Intelligent Mechatronic Systems, Human-Robot Interaction, Brain-Computer Interfaces, Mind-Controlled Systems, Medical Robotics, and Computer/Machine Vision.
Dr. Boshra Karimi Northern Kentucky University [biography]

Dr. Boshra Karimi is an Assistant Professor of Construction Management at Northern Kentucky University. She received the Ph.D. degree in Civil and Environmental Engineering (Construction Management) from Oklahoma State University, the MSc degree in Indust
Dr. Reza Abrishambaf http://orcid.org/0000-0002-6046-497X Miami University [biography]

Reza Abrishambaf received his PhD degree in Electrical and Electronic Engineering from Eastern Mediterranean University, Cyprus in 2012. His research interests include Distributed Control Systems, Intelligent Manufacturing, Wireless Sensor Networks, Indu

Download paper (695 KB)

Are you a researcher? Would you like to cite this paper? Visit the ASEE document repository at peer.asee.org for more tools and easy citations.

» Download paper

« View session

For those interested in:

Academia-Industry Connections
computer science
engineering
engineering technology
Faculty
information technology
professional
undergraduate