Integrating large language models with industrial simulation for multi-level decision support: an innovation management perspective in Industry 5.0

Purpose This research addresses the accessibility challenge in industrial simulation by integrating Large Language Models (LLMs) with simulation tools to democratize decision-making processes across organizational hierarchies in Industry 5.0 environments. Design/methodology/approach The study develops and implements an AI-driven interface that connects LLMs with industrial simulation models, enabling natural language interaction with complex simulation data. The methodology includes a comprehensive framework for data processing, query interpretation and result visualization. A real case study in the energy technology sector validates the approach through practical application in a manufacturing facility producing turbine components. Findings The integration successfully transformed simulation-based decision-making by enabling non-expert users to access and interpret complex simulation data through natural language queries. The case study demonstrated improved operational efficiency through better resource allocation and reduced decision-making bottlenecks. The system's validation confirmed the accurate interpretation of user queries and precise analysis of simulation data, supporting more inclusive and informed decision-making processes across organizational levels. Originality/value This research presents a comprehensive interface for integrating LLMs with industrial simulation models, introducing a novel approach to democratizing simulation-driven insights from an innovation management perspective. The study contributes to organizational theory by demonstrating how natural language interfaces can bridge the technical knowledge gap between simulation experts and decision-makers, transforming traditional decision processes and advancing the human-centric decision management vision of Industry 5.0.