<p>This work primarily focuses on optimizing the retrieval time in a flow rack automated storage and retrieval system, using an artificial intelligence technique. The goal is to prioritize the product to be retrieved while minimizing its retrieval time. To achieve this, an approach based on reinforcement learning, using more specifically the Q-learning algorithm, was implemented. In this case, the agent learns to make optimal retrieval decisions while interacting with the environment which, in the present case, is the rack. From then on, the agent explores this environment, performs actions, and receives rewards in return. The results highlight a significant improvement in retrieval time and rewards in a static environment. However, in the face of environmental dynamism, standard Q-learning shows limitations, paving the way for the exploration of more suitable variants.</p>