计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (9): 226-230.DOI: 10.3778/j.issn.1002-8331.1512-0140

• 工程与应用 • 上一篇    下一篇

业务流程的时延预测队列挖掘方法

曹芮浩1,方贤文1,王晓悦1,刘祥伟2   

  1. 1.安徽理工大学 信息科学系,安徽 淮南 232001
    2.安徽理工大学 经管学院,安徽 淮南 232001
  • 出版日期:2017-05-01 发布日期:2017-05-15

Delay prediction of queue mining method for business process

CAO Ruihao1, FANG Xianwen1, WANG Xiaoyue1, LIU Xiangwei2   

  1. 1.Department of Information Science,Anhui University of Science and Technology, Huainan, Anhui 232001, China
    2.College of Economic and Management, Anhui University of Science and Technology, Huainan, Anhui 232001, China
  • Online:2017-05-01 Published:2017-05-15

摘要: 过程挖掘是针对流程信息系统所记录下的日志进行分析,将业务流程真实过程还原的技术。目前已有的方法多是基于控制流与数据流的观点,针对任务运行状态的,无时延的业务过程进行挖掘。但在挖掘存在多任务的有时延的业务进程方面,目前的方法存在一定局限性。提出基于队列挖掘优化过程模型的方法,首先利用现有的基于过程挖掘的方法,挖掘业务流程的初始模型。再运用队列挖掘的观点对特定的顾客进行时延预测,挖掘出顾客的行为信息,以此对初始流程模型进行优化。最后通过实例验证了所提出的优化挖掘方法的有效性,优化后的流程模型不仅对事件日志有很好的重放效果,并且能够反应出多类别的,且存在时延的业务流程中任务的行为信息。

关键词: 队列挖掘, 时延预测, 服务日志, 过程挖掘, 行为轮廓

Abstract: Process mining is to analyze the log of the process information system, and to restore the real process of the business process. At present, the existing methods are based on the control flow and data flow, work for business process mining without time delay of activities and mining the process execution state. However, there are some limitations in the existing methods for the development of the time delay of multi tasks. This paper proposes a method for optimizing the process model based on queue mining. Firstly, it uses the existing process mining method, and finds the initial model of the business process. Then, by using the view of queue mining, it predicts the time delay of a target-customer to generalize the customer’s behavior information which is used to optimize the initial process model. Finally, the effectiveness of the proposed method is verified by an example. The optimized process model not only has a good replay on the event log, but also can reflect the behavior information of the task in the business process with time delay.

Key words: queue mining, delay prediction, service log, process mining, behavioral profile