Integrating operational and tactical decision-making in spare part inventory management

Tactical decisions in spare part inventory management (i.e. base stock levels) are taken based on aggregated demand and product data. On an operational level, interventions are taken to prevent and solve backorders. This research aims to integrate both levels of decision-making. In other words, we propose a method that considers the operational flexbility enabled by interventions when determining base stock levels. We use Deep Reinforcement Learning as our modeling approach.