Energy-efficient coordinated scheduling of stacker cranes and automated guided vehicles in a smart-meter automated storage and retrieval system

Abstract

Automated storage and retrieval systems require coordinated scheduling of stacker cranes and automated guided vehicles because their operations are coupled through task precedence, shared buffers, resource conflicts, and energy-related travel costs. This study examines a smart-meter warehouse operated by the Yunnan Power Grid Metrology Center and develops a mixed-integer linear programming model for integrated storage, retrieval, and transport scheduling. The model combines AGV travel-related energy, stacker-crane routing cost, and buffer dwell time in a weighted objective subject to capacity, timing, priority, and conflict constraints. An improved whale optimization algorithm is developed using chaos-based initialization, golden-sine-guided step updating, and adaptive opposition-based learning. Across 23 benchmark functions, the proposed algorithm achieved the best mean rank, and the overall algorithmic difference was significant according to a Friedman test. Pairwise Wilcoxon tests based on benchmark-function means also favored the proposed algorithm after Holm adjustment. In the practical case study, IWOA achieved the lowest mean final weighted objective in S1 and S3 and the lowest standard deviation in all three scenarios, while PSO obtained the lowest objective value in S2. The framework provides a model-based approach for coordinated scheduling in automated warehouses.