Previous research has clearly and consistently shown that flow time advantages accrue from splitting production lots into smaller transfer batches or sub-lots. Less extensively discussed, and certainly undesired, is t...Previous research has clearly and consistently shown that flow time advantages accrue from splitting production lots into smaller transfer batches or sub-lots. Less extensively discussed, and certainly undesired, is the fact that lot splitting may dramatically increase the number of setups required, making it impractical in some settings. This paper describes and demonstrates a primary cause of these “extra” setups. It then proposes and evaluates decision rules which selectively invoke lot splitting in an attempt to avoid extra setups. For the closed job shop environment tested, our results indicate that conditional logic can achieve a substantial portion of lot splitting’s flow time improvement while avoiding the vast majority of the additional setups which would be caused by previously tested lot splitting schemes.展开更多
The problem of minimizing the maximum lateness on a single machine with family setups is considered.To solve the problem,dominance property is studied and then introduced into the tabu search(TS)algorithm.With the dom...The problem of minimizing the maximum lateness on a single machine with family setups is considered.To solve the problem,dominance property is studied and then introduced into the tabu search(TS)algorithm.With the dominance property,most unpromising neighbors can be excluded from the neighborhood,which makes the search process always focus on the most promising areas of the solution space.The proposed algorithms are tested both on the randomly generated problems and on the real-life problems.Computational results show that the proposed TS algorithm outperforms the best existing algorithm and can solve the real-life problems in about 1.3 on average.展开更多
文摘Previous research has clearly and consistently shown that flow time advantages accrue from splitting production lots into smaller transfer batches or sub-lots. Less extensively discussed, and certainly undesired, is the fact that lot splitting may dramatically increase the number of setups required, making it impractical in some settings. This paper describes and demonstrates a primary cause of these “extra” setups. It then proposes and evaluates decision rules which selectively invoke lot splitting in an attempt to avoid extra setups. For the closed job shop environment tested, our results indicate that conditional logic can achieve a substantial portion of lot splitting’s flow time improvement while avoiding the vast majority of the additional setups which would be caused by previously tested lot splitting schemes.
基金supported by the Major State Basic Research Development Program of China(973 Program)(2002CB312205)the National Natural Science Foundation of China(60574077+2 种基金6087407160834004)the National High Technology Research and Development Program of China(863 Program)(2007AA04Z102)
文摘The problem of minimizing the maximum lateness on a single machine with family setups is considered.To solve the problem,dominance property is studied and then introduced into the tabu search(TS)algorithm.With the dominance property,most unpromising neighbors can be excluded from the neighborhood,which makes the search process always focus on the most promising areas of the solution space.The proposed algorithms are tested both on the randomly generated problems and on the real-life problems.Computational results show that the proposed TS algorithm outperforms the best existing algorithm and can solve the real-life problems in about 1.3 on average.
