Human-machine Interaction: Adapted Safety Assistance in Mentality Using Hidden Markov Chain and Petri Net

Abstract

This study proposes a cognition-adaptive approach for the administrative control of human-machine safety interaction through Internet of Things (IoT) data. As part of Industry 4.0, a human operator possesses various characteristics, but cannot be consistently understood as well as a machine. Thus, human-machine interaction plays an important role. This study focuses on incumbent challenges on the basis of estimated mental states. Given the operation logs from data recording hardware, a Hidden Markov model on top of a human cognitive model was trained to capture a production line worker’s sequential faults. Our study found that retaining workers’ attention is insufficient and tracking the state of perception is key to accomplishing production tasks. A safe workflow policy requires attention and perception. Accordingly, our proposed Petri Net enhances operation safety and improves production efficiency.