Buffer size decision for the flexible transfer line with rework paths using genetic algorithms

The buffer size decision for the flexible transfer line (FTL) gains more and more importance because of growing FTL complexity and production costs. The buffer size in front of the bay of machines in the FTL is still one of the major optimization problems faced by production engineers. In this paper, we attempt to find the near optimal buffer size for flexible transfer line with rework paths (FTLRP) that achieves the best throughput of the FTLRP. A genetic algorithm (GA) was applied to find buffer size of FTLRP. For the performance evaluation of the FTL, a novel aggregation technique proposed by Jingshan (2004), is used to find the throughput of the FTLRP at a given buffer size. In order to achieve the efficient use of the GA, multiple vectors distribution method (MVDM) is used for the genes arrangement. An application example was developed and after a number of operations based on GA, the sizes of all buffers for the FTLRP could be found.