Reducing cycle times in manufacturing and supply chains by input and service rate smoothing |
| |
Authors: | Sung-Seok Ko Richard Serfozo Appa Iyer Sivakumar |
| |
Affiliation: | a School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, GA, USAb School of Mechanical and Production Engineering, Nanyang Technology University, Singapore |
| |
Abstract: | A key performance parameter of a manufacturing network or supply chain is its cycle time; the time that a typical item spends in the network. A previous simulation study on a semiconductor assembly and test facility showed that cycle times could be reduced by having smooth input and service rates. This suggested that there is a “cycle time principle” that, for a system with a specified throughput or input rate, the shortest cycle times are obtained when the input and service rates do not vary over time. We prove that this principle is true for the M/G/1 and M/M/s queueing systems and Jackson networks. The analysis involves establishing several results on the concavity of waiting time probabilities and the convexity of expected waiting times and queue lengths, as functions of input and service rates. These results also have natural uses in other optimization problems. |
| |
Keywords: | |
本文献已被 InformaWorld 等数据库收录! |
|