Performance Enhancement of Flow Shop Scheduling Using Data Mining

Data Mining aims at discovering knowledge consisting of rules describing properties of data. In this paper, the optimized schedules for a flow shop scheduling problem are generated by the metaheuristic methods. The optimal schedules are used to generate the training set by identifying the predictor attributes affecting the schedules. The classifier model is generated based on decision tree induction by discretising the data in the training set based on chi2 algorithm. The classifier model is tested by applying for the new set of problems. The optimized schedule is generated by the classifier model is comparatively equal to that of the scheduling done by the metaheuristic methods. The sees tool is used for performing the data mining classification. Thus the rules generated by the Data Mining classifier allows the production managers to easily take decisions regarding the flow shop scheduling based on various objective functions and constraints.