A Novel Framework for Automated Warehouse Layout Generation

TL;DR

This paper introduces an AI-driven framework that automates warehouse layout generation by optimizing spatial arrangements to improve efficiency, using constrained search and feasibility verification to produce practical, optimal designs.

Contribution

The novel framework combines constrained beam search with feasibility verification to automatically generate and evaluate warehouse layouts, enhancing design speed and quality.

Findings

Successfully generates feasible warehouse layouts for various dimensions and shapes.

Produces optimal layouts considering accessibility, clearances, and connectivity.

Facilitates rapid exploration and selection of warehouse designs.

Abstract

Optimizing warehouse layouts is crucial due to its significant impact on efficiency and productivity. We present an AI-driven framework for automated warehouse layout generation. This framework employs constrained beam search to derive optimal layouts within given spatial parameters, adhering to all functional requirements. The feasibility of the generated layouts is verified based on criteria such as item accessibility, required minimum clearances, and aisle connectivity. A scoring function is then used to evaluate the feasible layouts considering the number of storage locations, access points, and accessibility costs. We demonstrate our method's ability to produce feasible, optimal layouts for a variety of warehouse dimensions and shapes, diverse door placements, and interconnections. This approach, currently being prepared for deployment, will enable human designers to rapidly explore and confirm options, facilitating the selection of the most appropriate layout for their use-case.