Abstract: The change mining in process models,as an important aspect of business process management,is one of the key points in ensuring the trustworthy management of business systems.The existing studies mainly focus on change analysis of control flow or fusion analysis between control flow and data flow in the business process,and less consider the impact of data changes on the change chain of behaviors and data in cross-domain interaction processes.To address this issue,a change analysis method based on Decision Model and Notation (DMN) was proposed for data impact interaction processes.The data-centric BPMN decision pattern was extracted to model the process context data in a separate decision model.Then,BPMN and DMN were integrated to model the fusion of control flow and data flow,and the soundness of the generated data-aware process model was analyzed.On this basis,the data impact analysis was used to identify the impact of data changes on subsequent process execution and linked data changes,so that the change propagation area under control flow and data flow constraints could be analyzed comprehensively.Simulations were performed by using CPN tools and evaluated in comparative experiments.The experimental results showed that the consistency of the propagation of interactive process changes under data influence was relatively high,and the model structure was optimized.

Key words: change-aware, decision model and notation, data impact, data-aware soundness, cross-domain interaction

摘要： 流程模型的变化挖掘作为业务流程管理的重要内容,是解决业务系统可信管理的关键点之一。已有研究主要是针对业务流程的控制流进行变化分析,或将控制流与数据流进行融合分析,较少考虑数据变化对跨域交互流程中行为和数据的变化链影响。基于此,提出了一种基于决策模型和标记（DMN）的数据影响交互流程的变化分析方法,首先提取以数据为中心的业务过程模型和标记（BPMN）决策模式,将流程上下文数据建模在单独的决策模型中。然后将BPMN与DMN相结合进行控制流与数据流融合建模,并分析生成的数据感知流程模型的合理性。在此基础上,通过数据影响分析识别数据变化对后续的流程执行及数据的联动变化的影响,从而综合分析控制流与数据流约束下的变化传播区域。最后利用CPN tools进行仿真模拟并通过比较实验进行评估,实验结果表明数据影响下的交互流程变化传播的一致性度相对较高,并优化了模型结构。

关键词: 变化感知, 决策模型和标记, 数据影响, 数据感知合理性, 跨域交互