In the wake of Industry 4.0 and the rise in the automation of material handling systems, overhead cranes have become very significant in the storage of steel in warehouses, as they are the link between production and processing. This thesis deals with the development of an AI assisted scheduling system for driverless overhead cranes operating within steel warehouses, which serve as a crucial interface between upstream steel production and its downstream processing stages. In metallurgical plants, smooth crane operation is highly critical for ensuring continuity in material flow, reducing idle times to a minimum, and maintaining stable production rates. The research addresses the modeling and optimization of a dynamic scheduling problem in which cranes must remove and reposition stacked steel materials under realistic operational constraints: precedence rules that require upper layers to be removed before lower ones, restricted movement directions along predefined rails or axes, and the sequential execution of lifting operations so as to achieve minimum total handling time.