| This dissertation shows that adding the capability of Adaptive Decision-Making Frameworks (ADMF) to a multi-agent system can result in significantly improved system performance across run-time situation changes. Specifically, ADMF can result in improved and more robust performance compared to the use of a single static decision-making framework. ADMF is an agent capability developed by this research that allows agents to adapt the decision-making frameworks in which they participate to fit their current situation. A decision-making framework (DMF) identifies a set of agents and specifies the set of interactions exercised by these agents as they determine how a goal or set of goals should be achieved. This research empirically shows that, for at least one multi-agent system, there is no one best decision-making framework for multiple agents across run-time situational changes. Further, this research justifies the implementation of ADMF by showing that allowing adaptation of decision-making frameworks to run-time variations in situation can result in improved overall system performance compared to static or random decision-making frameworks. |
