Job Shop Scheduling with Genetic Algorithms
Proceedings of the 1st International Conference on Genetic Algorithms
Knowledge Migration Based Multi-population Cultural Algorithm
ICNC '09 Proceedings of the 2009 Fifth International Conference on Natural Computation - Volume 04
Hi-index | 0.00 |
In this article, a multiagent system is proposed to solve Job Shop Scheduling Problems. In the proposed system, a number of autonomous agents cooperate in a Multi-Population Cultural Algorithm (MP-CA) framework. The proposed multiagent system consists of a number of groups of agents called sub-populations. The agents in each sub-population are co-evolving using a local CA. The local CAs are working in parallel and communicating to each other to exchange their extracted knowledge. The knowledge is migrated in the form of structured belief which is defined as a statistical records of an agent or a group of agents. Experiments show that our method outperforms some existing methods by offering better solutions as well as a better convergence rate.