A reinforcement learning model for material handling task assignment and route planning in dynamic production logistics environment

Abstract

The study analyzes the application of reinforcement learning (RL) for material handling tasks in Smart Production Logistics (SPL). It presents two contributions based on empirical results of a RL model in dynamic production logistics environment from the automotive industry. Firstly, an architecture integrating the use of RL in SPL. Secondly, the study defines various elements of RL (environment, value, state, reward, and policy) relevant for training and validating models in SPL. The study provides novel insight essential for manufacturing managers and extends current understanding related to research combining artificial intelligence and SPL, granting manufacturing companies a unique competitive advantage.