Enhancing Operator Engagement during AI-assisted Manufacturing Work Using Optimal State Deviation Feedback System

Abstract

The integration of Artificial Intelligence (AI) in manufacturing is shifting the focus of operators from manual labor to cognitive supervision roles. While this transition demands more engagement from operators, the less stimulating nature of monitoring tasks has, paradoxically, reduced operator involvement, consequently presenting new challenges in performance maintenance. Addressing this issue, our research adopted an iterative design science methodology to create a biocybernetic system that aims to enhance operator engagement in their evolving workplace. This system leverages physiological signals to intuitively display how much an operator’s engagement level deviates from an ideal state, ensuring operators stay aware of their psychophysiological state of engagement and can quickly adjust to any decreases in engagement. In this paper, we detail the 4-step process that led to the development of the first version of the system. Capitalizing on the physiological differences observed in manufacturing operators during “high” and “low” engagement scenarios, we defined a task-specific Optimal State Deviation Index (OSDI) formula. This formula enabled us to predict participants' engagement states with an 80.95 % success rate in our testing dataset.