Abstract
The flow path design problem is one of the most important issues of an automated guided vehicle (AGV) system design and control. A revised electromagnetism-like mechanism is proposed to solve the flow path design problem of unidirectional AGV systems under a reconfigurable manufacturing system environment in this paper. First, the formal model considering both loaded and empty travels is given. Then the preprocessing procedure for the guidepath network is proposed to improve the feasibility of solution and to reduce the computational complexity. In the revised electromagnetism-like mechanism, particles are encoded discretely, and Hamming distance based reduced variable neighbourhood search is adopted in the local search procedure. To reduce computation, only the best particle is searched in the local search procedure. In the moving procedure, each element of a particle is moved according to the ordering probability strategy. To verify the efficiency of the proposed algorithm, several computational experiments are carried out. The computational results show that the proposed algorithm is efficient, which is able to obtain optimal solutions to small scale problems and near optimal ones to large scale problems within reasonable computation time.
Acknowledgements
The work reported in this paper is supported by the National Basic Research Program (973Program) of China under Grant No: 2002CB312204. The authors would like to thank the anonymous reviewers for their valuable suggestions.