Trust based multi-agent cooperative load balancing system (TCLBS)

This paper proposes a trust based multi-agent cooperative load balancing system. The main functionalities are performed by a different agents. Load balancing agent (LBAgent), processing agent (ProcAgent), fetching agent (FetchAgent), and trust agent (TRAgent) use a centralized coordination architecture. We have implemented and tested TCLBS when there are both trusted and malicious agents. The results reveal the role of the trust agent ensures more reliable and effective service delivery depending on the behaviour of the agent. Experimental results reveal that the proposed trust computation method is apt in the presence of malicious agents to select the reliable agents to perform the tasks with load balancing.