Intelligent Warehousing based on numerical heuristic planning

The development of intelligent warehousing requires solutions for several challenges, including efficient allocation of goods, optimizing picking paths, improving overall turnover rates, and maximizing warehouse space utilization. While combinatorial optimization strategies and deep reinforcement learning have been used to address individual or multiple problems, there is a lack of a comprehensive planning and execution strategy for the warehousing scene itself. To overcome this challenge, this paper proposes utilizing AI planning technology in the warehousing field. We model warehouse scheduling as a planning problem. The proposed solution employs a state-of-the-art numerical heuristic planner to address the problems of warehousing model. Experimental results demonstrate that this approach can effectively simulate the warehousing scenario and provide feasible and highly interpretable solutions. This highlights the untapped potential of AI planning technology in the warehousing field.