Artificial Intelligence and Multi-Agent Approaches for Decision-Making and Business Process Optimization in Industrial Systems
Abstract
This study develops a conceptual and simulation-based framework for intelligent decision-making and process optimization in mineral processing systems. The framework integrates artificial intelligence, multi-agent coordination, adaptive control, and digital twins to support real-time optimization under uncertain and variable operating conditions. Systems analysis, mathematical modeling, simulation, and industrial data-processing methods are combined, while two-sample t-statistics assess differences between baseline and proposed scenarios across 50 independent simulation runs. The proposed architecture coordinates technological units, adjusts equipment parameters, and applies predictive analytics to decision support. Simulation results indicate improvements in resource utilization, energy efficiency, production stability, and product quality. By using production monitoring data and digital-twin representations, the framework supports industrial digital transformation without additional laboratory experiments and provides a basis for intelligent control and business-process optimization in mining and metallurgical operations.