By Stefanie Koperniak
Recent SDM grad Arman Shantayev, along with colleagues from MIT and Tecnológico de Monterrey, published a paper in the June 2026 edition of The International Journal of Advanced Manufacturing Technology about how augmented reality (AR) might help workers perform tasks in an industrial setting.
This paper draws from Shantayev’s SDM thesis research, developed with advisor Brian Anthony (principal research scientist in the Department of Mechanical Engineering and director of the MIT Master of Engineering in Manufacturing Program). Anthony leads a collaboration with Tecnológico de Monterrey utilizing low-cost, desktop fiber extrusion devices (FrEDs). Shantayev became interested in AR while working in Anthony’s lab in MechE, and realized that the devices could provide an opportunity to analyze how AR might affect or enhance how human operators are able to complete tasks. For his research, Shantayev built an AR application, ran a study of human subjects using the application to assemble FrEDs, and evaluated how AR affected their completion of the task.
“I hadn’t worked directly on AR before that, and I thought it would be an interesting skill to develop,” says Shantayev. “I was especially drawn to learning how to build AR applications and tools that could potentially help the operator on a manufacturing floor.”
Although relatively new to AR/VR, Shantayev arrived at SDM with substantial experience in engineering and operations—particularly oil and gas operations—working across production systems, field automation, and operational performance improvement. He earned an undergraduate degree in petroleum engineering from the University of Texas at Austin, and then moved back to his home country of Kazakhstan, where he worked for approximately ten years in production engineering and production operations roles for a Chevron joint venture.
“As I evolved through my career, I started to find myself being responsible for executing projects, and I felt like I needed to further develop my skills and knowledge,” he says. “SDM’s curriculum is well-suited to my passion for building products and delivering projects.”
Shantayev says that systems thinking was critical throughout the AR research, which involved considering the best way to break down complex tasks into simpler tasks, thinking of the entire device in terms of its individual parts, and iterating on creating the best user experience for the people using the AR application. Some participants assembled the device using the AR app while others used a conventional procedure manual. The researchers then compared the two groups to evaluate if there were any advantages or disadvantages of incorporating AR into the process. The research indicated that, overall, there was a substantial difference between the two groups, with both a faster decision-making time for the group using AR and a reduction in the number of errors. The research also tracked how much AR helped, or didn’t help, for the assembly of each component.
“For certain components that have a high level of ambiguity in the orientation, space, or shape of the object, people might find it harder to understand it using standard instructions on paper compared to using AR. However, for some easier-to-understand, more intuitive components, that might be different,” says Shantayev. “When we consider whether AR should be used more often on the manufacturing floor, the answer is that it depends on the complexity of the assembly task. AR may provide value for complex tasks, where spatial information is difficult to convey through conventional instructions, but that benefit has to be weighed against the cost and effort required to implement the technology.”
Shantayev is now drawing on both his professional background and his experiences at SDM to explore a startup idea focused on applying advanced AI-based process control methods to manufacturing and industrial production systems. The goal is to improve how complex industrial processes are controlled, helping operations respond more effectively to changing conditions and maintain stable, efficient production.
“In many continuous manufacturing processes, reducing unnecessary process variability can be important for maintaining stable and efficient operations,” he says. “We are exploring how machine-learning-based process control methods can improve upon conventional approaches. Our goal is to provide a platform that learns from operational data and can predict process behavior and optimize control decisions in real time.”
To read Shantayev’s paper, visit Springer Nature.