Journal of Xi'an Jiaotong University

Identification of Process Improvement Opportunity Based on Frequent Activity Sequential Pattern Mining Technology
Chen Liang¹£¬Gao Jianmin¹£¬Li Qing²£¬Chen Kun¹
£¨1.State Key Laboratory for Manufacturing System Engineering£¬Xi'an Jiaotong University£¬Xi'an 710049£¬ China£»2.School of Mechanical Engineering £¦ Automation£¬Beihang University£¬Beijing 100083£¬China£©

Abstract£ºTo guide the continuous improvement of quality management process by analyzing history workflow instance logs£¬ a method for identifying frequent activity sequential pattern related with given quality target is presented£®Aiming at the multiple loop structures in quality management process network£¬an improved graph pattern mining approach based on Apriori algorithm is put forward£¬ where a frequent pattern of size k£¬denoted as G_i£¬is joined with another frequent pattern of size k£¬denoted as G_j£¬to form a candidate frequent pattern of size (ª«k⁺¹)£¬if there exists a source vertex sª«in G_i and a sink vertex eª«in G_jsuch that the graph derived from G_i by subtracting node sª«equals with the graph derived from G_j by subtracting node e £®This coalition operator only considers the source vertexes and sink vertexes£¬which reduces the amount of redundant frequent candidates generated in the traditional Apriori iterations£®The external audit treatment process is used as a case study to illustrate the operation of the proposed method£¬and the result shows the effectiveness to discover the frequent activity sequential patterns to lead to the improvement of quality management process£®
Keywords£ºquality management process£»improvement opportunity£»workflow£» frequent activity sequential pattern