Optimal Repairman Assignment in Multi-level Repair Network for Weapon System Based on Multi-agent Simulation

Abstract

Considering the complicated interactions among repairmen and spare parts in multi-level repair network, a multi-agent simulation (MAS) model is built to evaluate the system operational readiness rate (ORR) under certain repairman assignment. The MAS model treats all line replaceable units (LRUs), spare parts at each level, and repairmen as agents, which have their own properties and action rules that are designed by state charts and events. By means of optimization via simulation (OvS) tool, OptQuest in the software AnyLogic 7, the maximum ORR and the optimal repairman assignment at each level for certain amount of total repairmen are obtained. The optimal repairman assignment is determined as the amount of total repairmen increasing to a value that the corresponding maximum ORR reaches the required ORR level. At last, a three-level repair network experiment is carried out and the results demonstrate that the proposed method is feasible and more reliable and accurate than existing methods.