Virtual/physical parameter space exploration for decision-making in production environments

This research proposes a simulation-based approach to exploring parameter spaces in production environments. The aim is to identify stable and reliable outcomes rather than solely optimising performance indicators. The proposed method overcomes the limitations of conventional simulations by incorporating "daydreaming factories" and foresighting to direct targeted exploration of the relevant parameter space regions. By visualising parameter sensitivities through surface plots, decision-makers can adjust simulation granularity in sensitive areas. A case study on injection moulding validates the approach, demonstrating its ability to support the identification of stable parameter combinations, informing decision-makers and closing the loop between strategy, simulation, and learning.