An Industrial HMI Temporal Adaptation based on Operator-Machine Interaction Sequence Similarity

The incorporation of Artificial Intelligence (AI) into Industrial Environments has brought about a Smart Industry revolution, improving efficiency and simplifying complex industrial processes. However, these technological advances remain primarily focused on the process, and pay little attention to industrial Human-Machine Interfaces (HMI), the bridge between the operator and the industrial process.Current industrial HMIs have a static design, and are focused exclusively on the control and visualization of process information. They fail to take into account user behaviour and skills, information key to understanding how the operator interacts with the production process. Thus, the potential beneficial outcomes of considering operator-machine interaction in terms of efficiency and productivity, make a compelling case for industrial HMIs that can adapt to different operators based on their skills and process knowledge.This paper proposes a Machine Learning (ML) based method-ology capable of analysing operator-machine interaction and detecting the variability of interaction patterns for repetitive similar sequences in monitoring and control tasks. The method-ology generates a set of adaptation rules that improve Usability and User Experience, and hence operator working performance. To validate the proposed methodology, an experiment with real operators was conducted.