Human–Machine Coordination in Adaptive Industrial Automation Systems

Abstract

Coordination of Human-Machine in Adaptive Industrial Automation refers to the integration of intelligent systems with human operatives to improve production efficiency and decision-making within industrial processes. The existing approaches are restricted by limited adaptability, contextual analysis, and inefficiency in human-machine interactions. This research presents an approach that uses Deep Learning (DL) based on Gannet Optimization-driven Adaptive Long Short-Term Memory (GO-ALSTM) to optimize the coordination of human-machines within adaptive industrial automation systems. Industrial operational data are subjected to preprocessing by Min–Max normalization and feature reduction using Principal Component Analysis (PCA), to enhance data quality, reduce redundancy, and improve the representation of underlying patterns for subsequent analysis. The GO-ALSTM approach makes use of ALSTM for identifying long-term dependencies and GO for optimizing parameters. The research makes use of deep learning libraries within the Python programming environment. The experimental evaluation proves that the proposed technique achieved precision, F1-score, recall, and accuracy 96.8 ± 0.4%, 97.0 ± 0.4%, 97.2 ± 0.5%, and 97.0 ± 0.3%,  respectively, surpassing both the Deep Learning (DL) and Machine learning (ML) methods. Proposed model is able to significantly enhance the adaptability, reliability, coordination capability, and intelligent decision-making process in industry operations with complex and changing conditions.