A Framework for Performance Control ofDistributed Autonomous Agents

摘要

We propose an autonomous and scalable queueing theory-based methodology to control the performance of a hierarchical networkrnof distributed agents. Multi-agent systems (MAS) such as supply chains functioning in highly dynamic environmentsrnneed to achieve maximum overall utility during operation. Hence, the objective of the control framework is to identify therntrade-offs between quality and performance and adaptively choose the operational settings to posture the MAS for better utility.rnBy formulating the MAS as an open queueing network with multiple classes of traffic we evaluate the performance andrnsubsequently the utility, from which we identify the control alternative for a localized, multi-tier zone.