Construction workers face daily risks from extreme heat, especially during Arizona summers. Researchers at Arizona State University are using artificial intelligence to try to reduce that danger before it becomes a medical emergency.
The Safety Automation and Visualization Environment Laboratory, known as SAVE Lab, built a system that tracks wearable biosensors and environmental conditions on job sites. It sends workers personal alerts when signs of fatigue or heat stress appear, before symptoms start.
Siyuan Song, an associate professor of construction engineering who leads the lab, said the goal is to catch problems early. She said two workers in the same conditions can respond very differently to heat.
"It monitors conditions no human observer could track," Song said. The system shifts safety management from reacting to incidents toward preventing them, she added.
The technology was developed in partnership with Achen-Gardner Construction, Willmeng Construction and CHASSE Building Team. These companies tested the sensors directly on their job sites.
How AI Is Used Across Construction Projects
SAVE Lab researchers also build virtual reality training environments. These simulate hazards so workers can learn how to respond without real-world risk.
The lab uses large language models too. Researchers apply them to study accident reports and evaluate how well safety training programs work.
Shiva Pooladvand, an assistant professor of construction engineering, leads a separate research group focused on human-centered construction technology. Her team combines robotics, computer science and immersive environments.
"I use AI to analyze multimodal data captured from workers and the environment," Pooladvand said. She said this helps identify unsafe conditions and workers who may be at risk.
Ricardo Eiris, an assistant professor of construction management, studies how AI can improve training. His lab combines AI with virtual reality, digital twins and drones.
"I leverage AI to support immersive experiences, model human behavior and create more realistic, responsive systems," Eiris said.
Kenn Sullivan, a professor of construction management, uses AI for statistical modeling and to document research methods for future students.
These approaches were on display at the AI for Construction 2026 Contractors Summit in July. Industry leaders discussed how they use AI for business development, project controls and safety analysis.
Tim Becker, programs chair of the Del E. Webb School of Construction, said the summit grew out of requests from an executive council. Companies, faculty and students wanted a shared space to learn from one another.
Changes Happening In The Classroom
AI use in classrooms has grown fast. Over nine months, Sullivan said AI tools went from completing almost none of his students' assignments to completing up to 60% of them.
That shift led him to redesign his course. Students now build AI skills and working apps rather than relying only on traditional assignments.
In Pooladvand's class, students discuss real construction projects and learn to verify AI-generated information using their own engineering judgment.
Song teaches a course on construction safety that covers computer vision, wearable sensors and virtual reality. Case studies from the SAVE Lab give students hands-on experience with the technology.
Song said responsible AI use matters because construction mistakes carry physical consequences. An unverified AI output in a safety plan, she said, can become a hazard for real people on real job sites.
