The importance of production scheduling increases as changing customer demands increase its complexity. At the same time, current research often neglects practical constraints, e.g., changeovers or transportation. To address this issue, we develop a reinforcement learning application for production scheduling that considers practical constraints. In prior research, we derived requirements for a production scheduling application to be applicable in practice. In this publication, we implement the application and evaluate the performance in a real-world production scenario using a Design Science Research methodology. After explaining how the practical requirements are implemented in the scheduling application, we show that it creates valid production schedules that consider the identified practical constraints. Afterwards, in the evaluation, we find a performance improvement of up to 8% in comparison to priority rules. Our results guide the application of production scheduling research results in practice by providing a scheduling application to be used by other researchers and companies.