Ressourcenorientierte Auftragsregelung in einer hybriden Produktion mittels betriebsbegleitender Simulation

Description

Modern production systems must be capable of reacting to changing environments within minimum time. This requires a high degree of transparency and responsiveness, especially in production control, so that inefficiencies can at best be identified early and avoided proactively. In order to facilitate this even in complex, multi-stage manufacturing processes, a production-synchronous digital twin is required which is equipped with forecasting and optimization competencies.

This dissertation provides a holistic methodology for the efficient processing of manufacturing orders in a hybrid production environment, even in case of unforeseen deviations, by using simulation-based optimization methods during operation. This novel approach no longer controls production based on planning data and reacts with locally optimal solutions, but continuously adapts the system to an overall optimum by integrating realtime data. Discrete-event simulation methods at its core are enriched with runtimeefficient modelling for fluid-like bulk materials as well as a material flow control based on thermodynamic models in order to ensure a sufficient level of detail for use with short-term problems

Due to the integrated energy simulation, the real dependencies in the hardening process can be taken into account with sufficient accuracy in the material flow simulation. By adding specific optimization methods to this time-dynamic modeling, it is possible to automate the decision-making process during simulation. Finally, a real-time representation of the production is available at any time in the simulation model through a suitable coupling with decentrally provided production data. On this basis, the adherence to the production targets can be continuously monitored and the production plan can be ad-justed proactively when an intervention is required.

Comprehensive experimental investigations within the course of case studies in calcium silicate masonry plants ensure the validity and applicability of the developed methodology. Its generic design and the establishment of a building block library for the simulation of hybrid production processes allow an immediate transfer to other industrial sectors.