Reinforcement Learning (RL) is an optimization method from the field of Machine Learning. It is characterized by two interacting entities referred to as the agent and the environment. The goal of RL is to learn how an agent should act to achieve a maximum cumulative reward in the long-term. A Discrete Event Simulation Model (DESM) maps the temporal behavior of a dynamic system. The execution of a DESM is done via a simulator.
The concept of an Experimental Frame (EF) defines the general structure used to separate the DESM into the dynamic system, called the Model Under Study (MUS), and its application context. This supports the diverse use of a MUS in different experimental contexts. This paper explores the generalized integration of discrete event simu-lation and RL using the concept of EF. The introduced approach is illustrated by a case study that has been implemented using MATLAB/Simulink and the SimEvents blockset.