Data Mining Based Approach for Jobshop Scheduling

In manufacturing system, there usually have been some unpredictable dynamic events, which would make the production scheme invalid. Therefore, it’s necessary to inject some new vitality to traditional scheduling algorithms. To harness the power of complex real-world data in manufacturing processes, a jobshop scheduling algorithm basing on data mining technique is presented. This approach is explored in view of seeking knowledge that is assumed to be embedded in the historical production database. Under the proposed scheduling system framework, C4.5 program is used as a data mining algorithm for the induction of rule-set. A rule-based scheduling algorithm is elaborated on the basis of the elaborated data mining solutions. The objective is to explore the patterns in data generated by conventional intellectualized scheduling algorithm and hence to obtain a rule-set capable of approximating the efficient solutions in a dynamic job shop scheduling environment. Simulation results indicate the superiority of the suggested approach.