Composite Dispatching Rule Generation through Data Mining in a Simulated Job Shop


Baykasoglu A., Gocken M., ÖZBAKIR L. , KULLUK S.

2nd International Conference on Modelling, Computation and Optimization in Information Systems and Management Sciences, Metz, France, 8 - 10 September 2008, vol.14, pp.389-391 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 14
  • Doi Number: 10.1007/978-3-540-87477-5_42
  • City: Metz
  • Country: France
  • Page Numbers: pp.389-391

Abstract

In this paper, a new data mining tool which is called TACO-miner is used to determine composite Dispatching Rules (DR) under a given set of shop parameters (i.e., interarrival times, pre-shop pool length). The main purpose is to determine a set of composite DRs which are a combination of conventional DRs (i.e., FIFO, SPT). In or-der to achieve this, full factorial experiments are carried out to determine the effect of input parameters on predetermined performance measures. Afterwards, the data set which is obtained from the full factorial simulation analyses is feed into the TACO-miner in order to determine composite DRs. The preliminary verification study has shown that composite DRs have an acceptable performance.