A data mining-based solution method for ow shop scheduling problems

Scheduling is the process of determining where and when to perform manufacturing measures, which is required to conduct activities in a timely, ecient, and cost-eective manner. In this paper, an algorithm is proposed as a solution to the ow shop scheduling problem which holds an important place in the scheduling literature. The path relinking algorithm and data mining are used to solve the ow shop scheduling problem studied here. While DM is used for globally searching the solution space, path relinking is used for local search. Data mining is a method for extracting the embedded information in a cluster that includes implicit information. Path relinking is an algorithm that advances by making binary displacements in order to convert the initial solution to the guiding solution and it is repeated by assigning the best obtained solution within this process to the starting point. The eciency of the model for Taillard's ow shop scheduling problems was tested. Consequently, it is possible to solve the large-size problem without considerable mathematical background. The obtained results showed that the proposed method comparatively performed as good as other metaheuristic methods.