基于Petri网的业务流程模型配置优化方法

发布时间：2018-06-07 04:07

本文选题：Petri网 + 变化域　；参考：《安徽理工大学》2017年硕士论文

【摘要】：业务流程管理有很多优点,比如:提高企业运行效率,尽可能的缩短运行时间,通过灵活的运用业务流程管理解决各种变化问题。同一个流程可能对应不同的模型,但有的不符合要求,因此会产生模型变化。同时,业务流程的质量取决于模型的优劣,而业务流程挖掘伴随互联网时代快速发展,它是一种管理领域的较新应用,分析业务流程对应的日志,挖掘可能存在的隐变迁、阻塞变迁,获得更加完备的流程模型,同时提高模型的工作效率。所以结合配置变迁优化模型是目前企业管理的重点。目前对给定目标模型的变化域的分析大部分通过变化的活动节点搜索变化范围,计算过程繁琐;挖掘业务流程模型隐变迁主要基于业务流程序列进行分析,但很少考虑跨序列间的关系,因此对挖掘业务流程隐变迁有一定的影响;仅仅给定日志,没有目标模型时,常常通过静态定位变化区域,然后优化分析业务流程模型,完善已有的模型。本文结合Petri网的方法和行为轮廓的思想,通过流程树切挖掘业务流程模型隐变迁,以及考虑在给定目标模型或者没有目标模型的情况下,从不同角度对变化域进行优化分析。本文的主要贡献如下:针对给定目标模型,寻找目标模型变化域的问题,提出基于行为轮廓的交互行为,通过边界变迁和跨边界变迁进行优化,确定所有的可疑点,通过服从度和支持率确定目标模型的变化域,不仅简化计算,而且减少了分析变化域的难度。针对挖掘业务流程模型隐变迁,并进行优化的问题,给出流程树切挖掘业务流程模型隐变迁的方法。首先根据发生频数较高的日志序列得到初始模型,然后再根据流程树切预处理事件日志,找到可能变化的区域,挖掘可能存在的隐变迁,最后通过评价指标判定带隐变迁的模型是最优模型,说明模型中存在隐变迁。该方法不仅从业务流程序列进行分析,而且考虑跨序列间的关系,从多个角度对模型进行分析,克服了已有方法的局限性。对于业务流程模型的智能优化问题,结合Petri网的方法和行为轮廓的思想,从日志行为方面分析检测业务流程模型的变化域,首先利用日志与模型的匹配关系以及动态定位来确定模型中的变化域,然后考虑接口部分,通过配置变迁优化模型,最后利用行为轮廓一致性度分析判定配置后模型是最优模型,从而克服静态寻找变化区域的有限性。
[Abstract]:Business process management (BPM) has many advantages, such as improving the efficiency of enterprise operation, shortening the running time as much as possible, and solving all kinds of change problems through flexible application of BPM. The same process may correspond to different models, but some do not meet the requirements, resulting in model changes. At the same time, the quality of business process depends on the advantages and disadvantages of the model, and business process mining is a relatively new application in the field of management with the rapid development of the Internet era. Block transition, obtain a more complete process model, while improving the efficiency of the model. Therefore, the combination of configuration transition optimization model is the focus of enterprise management at present. At present, the analysis of the change domain of a given target model is mostly based on searching for the range of changes by changing active nodes, and the calculation process is cumbersome, while mining the hidden transition of the business process model is mainly based on the analysis of the business process sequence. However, the cross-sequence relationship is rarely considered, so it has a certain influence on mining business process implicit transition. When only given log and no target model, the change region is usually located statically, and then the business process model is optimized and analyzed. Perfect the existing model. This paper combines the method of Petri net and the idea of behavior contour, mining the hidden transition of business process model through process tree cutting, and considering the optimization analysis of change domain from different angles under given target model or no target model. The main contributions of this paper are as follows: aiming at the problem of finding the change domain of the target model for a given target model, an interactive behavior based on behavioral contour is proposed, which can be optimized by boundary transition and cross-boundary transition to determine all suspicious points. The change region of the target model is determined by the degree of serviceability and the approval rate, which not only simplifies the calculation, but also reduces the difficulty of analyzing the change domain. Aiming at the problem of mining hidden transition of business process model and optimizing it, a method of mining implicit transition of business process model based on process tree cutting is presented. First, the initial model is obtained according to the log sequence with high frequency of occurrence, then the event log is preprocessed according to the process tree to find the region that may change, and the potential hidden changes are mined. Finally, the model with hidden transition is determined to be the optimal model by evaluation index, which shows that there is implicit transition in the model. This method not only analyzes the business process sequence, but also considers the relationship between the different sequences, and analyzes the model from several angles, which overcomes the limitation of the existing methods. For the intelligent optimization of business process model, combined with the method of Petri net and the idea of behavior profile, the change domain of business process model is analyzed from log behavior. Firstly, the matching relation between log and model and dynamic location are used to determine the change region in the model, then the interface part is considered, and the transition optimization model is configured. Finally, the behavior contour consistency analysis is used to determine that the post-configuration model is the optimal model, which overcomes the limitation of static searching for the region of change.
【学位授予单位】：安徽理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F272;TP301.1

【相似文献】