A novel task allocation protocol in agent organizations and its application to health care problems

In this paper, a novel protocol to address task allocation in agent organizations is proposed. Compared with existing related works, this protocol takes the structure of agent organizations into account during task allocation processes, and allows not only neighboring agents but also indirectly linked agents in an organization to help with a task if needed. Through this way, agents can achieve more efficient and robust task allocations in loosely coupled distributed environments. A set of experiments are presented to evaluate the efficiency and scalability of the protocol. The experiment results demonstrate that the protocol can work efficiently in different situations. In addition, an example of its application to a health care problem is also introduced, which shows its potential application to some real cases.