%0 Journal Article %A WANG Jiahai %A CHEN Yu %T Data driven Job Shop production scheduling knowledge mining and optimization %D 2018 %R 10.3778/j.issn.1002-8331.1705-0139 %J Computer Engineering and Applications %P 264-270 %V 54 %N 1 %X During the evaluation process of the Job Shop scheduling problem, the knowledge hidden in mass data of the plant plays the vital role in conducting the production optimization. Based on the analysis of the mass data and specialized knowledge in scheduling domain, ontology-based relation model is established and the knowledge is represented. Considering the importance of the combination within the data mining and knowledge, the decision tree CART(Classification and Regression Tree) algorithm integrated with SVM(Support Vector Machine) is introduced to obtain the data-driven scheduling rule, an instance is laid out to show the process of rule acquisition under the scheduling knowledge mining framework. When it comes to the implementation, the knowledge combining with artificial fish swarm algorithm is applied to artificial fish initialization optimum design. Finally, a calculation analysis of samples is carried out and the results show the effectiveness and accuracy of the improved algorithm, with more practical solution and effectively reduce the makespan and enhance productivity. %U http://cea.ceaj.org/EN/10.3778/j.issn.1002-8331.1705-0139