Warrior Vehicle Fleet Sustainment Using Intelligent Agent Simulation
暂无分享,去创建一个
Abstract The Warrior Infantry Fighting Vehicle(IFV) is one of the key tracked combat vehicles in the UK Army. It was first introduced in 1988. A modernisation programme is currently underway to enhance 643 vehicles to serve to 2040 and beyond. The Warrior is typical of military assets that have to be acquired, maintained, supported and deployed. Effective materials and logistics support for the life of the asset is necessary to give the army a capability to defend the country and keep world peace. In the military world, events are uncertain. An asset can be used relatively lightly in peace time for training and readiness preparation. When it is deployed, it is used intensely and probably in situations not foreseen in the original design specification. Compared with a commercial vehicle that is designed to be continuously heavily used, military uncertainties make the planning for spares and repairs very difficult. Responding to the dynamics of military logistics, inventory planners have to make decisions on how many spares to order from the manufacturer and when, where to store the spares, and when to send them to the units. Maintenance decisions are also made by military engineers to pull vehicles into depots for scheduled maintenance, deal with unexpected repairs, and make sure all people, equipment and spares are coordinated for the maintenance work. Planning for the worst case scenarios provides enhanced resilience to military needs, but is likely to be unnecessarily costly. Compared with an inventory management problem that has steady demand and supply, optimal military logistics could be better served by adapting the behaviour of the planners to suit the dynamics of the deployment scenarios. This paper reports on a military logistics sustainment model built using an agent based simulation platform, with the Warrior vehicle fleet as the case study. The model proves an effective tool to help military planners evaluate different spares inventory policies to match deployment demands.
[1] Sean Luke,et al. Cooperative Multi-Agent Learning: The State of the Art , 2005, Autonomous Agents and Multi-Agent Systems.
[2] J. Banks,et al. Discrete-Event System Simulation , 1995 .
[3] Stéphane Galland,et al. A Verification by Abstraction Framework for organizational Multi-Agent Systems , 2008 .
[4] Richard E. Nance,et al. A history of discrete event simulation programming languages , 1993, HOPL-II.