Task scheduling (TS) and multi-agent-path-finding (MAPF) are two cruxes of pickup-and-delivery in automated warehouses. In this paper, the two cruxes are optimized simultaneously. Firstly, the system model, task model, and path model are established, respectively. Then, a task scheduling algorithm based on enhanced HEFT, a heuristic MAPF algorithm and a TS- MAPF algorithm are proposed to solve this combinatorial optimization problem. In EHEFT, a novel rank priority rule is used to determine task sequencing and task allocation. In MAPF algorithm, a CBS algorithm with priority rules is designed for path search. Subsequently, the TS-MAPF algorithm which combines EHEFT and MAPF is proposed. Finally, the proposed algorithms are tested separately against relevant typical algorithms at different scales. The experimental results indicate that the proposed algorithms exhibited good performance.

In this paper, we address a multiload AGVs workshop scheduling problem with limited buffer capacity. This has important theoretical research value and significance in the manufacturing field in considering the efficient multiload AGVs widely used today, and in the limited buffer area in production practice. To minimize the maximum completion time, an improved ant colony optimization-simulated annealing algorithm based on a multiattribute dispatching rule is proposed. First, we introduce a multiattribute dispatching rule, which combines two attributes, delivery completion time and input queue through dynamic weights that are determined by the information about the system, using the multiattribute dispatching rule to construct the initial solution. Then, with the ant colony optimization-simulated annealing algorithm as the basic framework, we propose a method for calculating transfer probability based on the multiattribute dispatching rule, which obtains heuristic information through the proposed rule. Further, we propose a path branch mechanism and dynamic equilibrium mechanism, aiming to efficiently construct the ant path and dynamically adjust ant path distribution. We propose a key job strategy and design a 2-opt neighborhood search method based on key jobs. Data experiments demonstrate the multiattribute dispatching rule is superior over other heuristic dispatching rules; the algorithm improvement strategies proposed are effective when used simultaneously or separately. Further, the proposed algorithm in this paper is superior over other heuristic algorithms and adapts to all kinds of instances.

The cross-aisles shuttle-based storage/retrieval system has not only storage function but also the sorting function and makes full use of warehouse space to achieve high-density storage. It uses “part-to-picker” order picking mode to respond quickly to orders and shorten sorting time. In this paper, through the analysis of the system, an effective evaluation method for the efficiency and time of system picking under a single instruction operation cycle is presented. The objective function of system minimum cost under the condition of satisfying customer’s demand is constructed. The objective function is solved by an improved particle swarm algorithm based on the optimized initial particle swarm optimization. By optimizing, we can find the optimal configuration of the system (i.e., the number of tiers, number of aisles and number of bays, number of picking stations, and number of lifts with minimal system cost). Finally, the impact of different configurations on system performance is summarized. This method can guide the design planner to design a more reasonable system under minimum cost control.