SAMBA: A reference framework for Human-in-the-Loop in adaptive Smart Manufacturing

In modern smart manufacturing environments, human involvement remains critical for addressing complex tasks that require adaptability and decision-making, despite the growing presence of automation and artificial intelligence. This paper introduces SAMBA - Service-Augmented Manufacturing-Based Approach, an innovative framework designed to optimize human-in-the-loop processes in smart manufacturing. The problem addressed involves the challenge of effectively integrating human operators with advanced automation technologies to reduce errors and increase process adaptability. The framework employs Large Language Models to extract procedural specifications from unstructured sources and convert them into structured instructions. By integrating these instructions with Extended Reality technologies to assist human operators, SAMBA aims to minimize errors and adapt processes to complex, dynamic production environments. The framework proposes the enhancement of Manufacturing Execution Systems by incorporating Artificial Intelligence techniques to enable adaptability and automatic exception correction. Finally, an adaptive orchestrator models each resource as a service accessible via Industrial APIs, promoting interoperability and integration within the production ecosystem.