Applying conflict management strategies in BDI agents for resource management in computational grids
ACSC '02 Proceedings of the twenty-fifth Australasian conference on Computer science - Volume 4
| Hi-index | 0.00 |
We propose an iterative broker approach for solving a multi-agent scheduling problem in an inherently distributed environment. The agents have incomplete information about the environment and incomplete models of other agents, and are reluctant to disclose any information to other agents. The algorithm allows the agents to reach a solution close to the global optimum found in the centralised approach. The algorithm is demonstrated through two scenarios. We have applied the algorithm to a case study and the result is as good as in the centralised approach. Experimental results on random data are also provided.