Towards Data Management and Data Science for Internal Logistics Systems using Process Mining and Discrete-Event Simulation

Abstract

Internal logistics systems are often planned with the assistance of simulation. However, with increasing digitization, there is also growing trend towards data-oriented tools such as data and process mining. These tools offer promising novel approaches, for instance for the detection of bottlenecks. At the same time, they require substantial amounts of process data, which real-world systems often cannot provide in sufficient quality. In this article, a methodology is developed that allows to combine process mining and simulation. The focus lies on minimizing the effort for data processing, and on obtaining and verifying contextually meaningful improvements. This methodology is subsequently applied to a practical example, which allows statements on its effort and usefulness to be made.