In most of Light Emitting Diode (LED) semiconductor fabricating line, AGVs (Automated Guided Vehicles) are used to transport the products between the facilities in order to ensure dust-free operation in the transportation systems. It has been demanded to realize the route planning system which speedily derives a feasible transportation route so that each AGV does not interfere with other AGVs. In this paper, we applied an agent-based optimization method for the route planning problems with multiple AGVs in LED fabricating bay in which the total transportation time is minimized. In the agent-based method, each AGV agent independently searches its candidate of route. The imitated pheromone is effectively used to generate the route which avoids the collision among AGVs. In order to evaluate the performance of the proposed method, the lower bound of the original problem is obtained by using the Lagrangian relaxation technique. The performances of the proposed method are evaluated by using the value of duality gap for a large scale problem. The effectiveness of the agent-based optimization method is demonstrated.