Dynamic task allocation strategies for AGVs in smart warehouse environments

The implementation of dynamic task assignment between AGVs is an important step in enhancing the overall efficiency of AGVs in a smart warehouse environment. However, existing zone-based systems limit flexibility in terms of AGVs assigned to zones, thus resulting in workload imbalance and increased AGV idle time. To mitigate these drawbacks of existing zone-based systems, a centralized proximity-based dynamic task allocation framework is proposed using CoppeliaSim for dynamic assignment of tasks to AGVs based on the Euclidean distance between each AGV and task location. Such a system avoids the assignment of AGVs to zones and provides adaptability in terms of redistribution of workload between AGVs. The experimental results show an increase in overall average utilization of AGVs by 118%, a reduction in overall idle time by 78.7%, and a reduction in variance of task completion time by 59.8% when compared to static allocation systems. Additionally, overall task completion time is improved by 3.5% when using dynamic allocation compared to static allocation systems. Hence, overall consistency in terms of system stability is achieved. The results of the paper prove that using a lightweight proximity-based system for AGV allocation in a warehouse is not only effective but also scalable.