Generalizable and Understandable Self-Learning Approaches for Dynamic, Large-Scale Resource Management Optimization Problems

The goal of GULAD is to develop data-driven solutions that address complex optimization problems in manufacturing. Unlike traditional methods that rely heavily on expert knowledge and domain-specific approaches, GULAD leverages advanced AI techniques—such as deep learning and deep reinforcement learning—to create adaptive and scalable solution methods. These methods (1) continuously improve and adapt in uncertain and dynamic environments and (2) generalize effectively to large-scale problems and their variations. We apply these AI-driven techniques to tackle machine scheduling problems faced by Goodyear, aiming to optimize efficiency and performance in real-world manufacturing scenarios.