Tactical analysis of manufacturing processes via simulation

This thesis aims to make tactical analysis of manufacturing processes more accessible and effective through simulation. It addresses the challenge of creating simulation models that are both easy to build and sufficiently expressive to represent realistic production dynamics. The focus is on enabling analysts with limited programming experience to evaluate manufacturing systems at the tactical level, including resource allocation, process reconfiguration, and throughput optimization. The research proposes a hybrid activity- and resource-based simulation paradigm that combines the intuitive abstraction of Business Process Modeling with the operational detail of manufacturing simulation. This approach bridges conceptual modeling and executable simulation, allowing users to construct and analyze models through structured, model-driven specifications rather than code. The framework supports stochastic behavior, resource constraints, and inter-process dependencies, ensuring both usability and representational fidelity. A comprehensive methodology is introduced, covering model definition, data-driven parameterization, and systematic experimentation to support tactical decision-making. Case studies show how the simulation environment enables exploration of alternative process configurations and performance trade-offs without requiring specialized programming skills. The results demonstrate that simulation-based tactical analysis helps connect operational detail with strategic intent in manufacturing. By supporting what-if analyses under uncertainty and variability, the framework transforms simulation into a practical decision-support tool and promotes experimentation and continuous improvement. Beyond its immediate applications, the research establishes the foundation for a multi-model platform integrating data from multiple sources and levels of abstraction. Future extensions include linking forecasting and logistics data to align market demand with supply-chain constraints, incorporating transactional processes to capture information flow and decision delays, and integrating shop-floor sensor and IoT data for dynamic model and KPI updates. These directions support progressive model fidelity, component reuse, and consistent semantics, positioning the platform as a basis for data-centric manufacturing planning and future digital-twin ecosystems.