A Model Repair Approach Based on Petri Nets by Constructing Free-Loop Structures

He, Zhaohan; Du, Yuyue; Qi, Liang; Du, Haisen

doi:10.1109/access.2019.2899963

Cited by 4 publications

(5 citation statements)

References 39 publications

(48 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, the predictive quality might be able to be improved by incorporating quality performance measures of process discovery algorithms apart from fitness scores to investigate the impact of the process model quality on the proposed approach. Additionally, repair methods might be beneficial to increase the process quality and may have a positive impact on the predictive performance [74]- [76]. The outcome of such studies might further increase the metric scores on the next event prediction that we have reported.…”

Section: Discussionmentioning

confidence: 83%

Decay Replay Mining to Predict Next Process Events

Theis

Darabi

2019

IEEE Access

View full text Add to dashboard Cite

In complex processes, various events can happen in different sequences. The prediction of the next event given an a-priori process state is of importance in such processes. Recent methods have proposed deep learning techniques such as recurrent neural networks, developed on raw event logs, to predict the next event from a process state. However, such deep learning models by themselves lack a clear representation of the process states. At the same time, recent methods have neglected the time feature of event instances. In this paper, we take advantage of Petri nets as a powerful tool in modeling complex process behaviors considering time as an elemental variable. We propose an approach which starts from a Petri net process model constructed by a process mining algorithm. We enhance the Petri net model with time decay functions to create continuous process state samples. Finally, we use these samples in combination with discrete token movement counters and Petri net markings to train a deep learning model that predicts the next event. We demonstrate significant performance improvements and outperform the state-of-the-art methods on nine real-world benchmark event logs.

show abstract

Section: Discussionmentioning

confidence: 83%

Decay Replay Mining to Predict Next Process Events

Theis

Darabi

2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The approaches in [33][34][35][36][37][38][39][40][41][42][43][44][45] use the concept of Logic Petri net to repair a process model represented as a Petri net. They differ mainly by the conformance checking technique used (e.g., alignment-based [34], token-replay-based [35] and footprint-based [36]), requirements for the input Petri net (e.g., concurrency structure [37] or non-free choice structures [43]), and proposal repair (e.g., building choice structure [36] or loop structure [44]). In all the approaches, the authors argue that by only changing the structure in the Petri net associated with the deviations the initial model is minimally changed.…”

Section: Analysis Of Literature On Existing Process Model Repair Tech...mentioning

confidence: 99%

On the use of domain knowledge for process model repair

Revoredo

2022

Softw Syst Model

View full text Add to dashboard Cite

Process models are important for supporting organizations in documenting, understanding and monitoring their business. When these process models become outdated, they need to be revised to accurately describe the new status quo of the processes in the organization. Process model repair techniques help at automatically revising the existing model from behavior traced in event logs. So far, such techniques have focused on identifying which parts of the model to change and how to change them, but they do not use knowledge from practitioners to inform the revision. As a consequence, fragments of the model may change in a way that defies existing regulations or represents outdated information that was wrongly considered from the event log. This paper uses concepts from theory revision to provide formal foundations for process model repair that exploits domain knowledge. Specifically, it conceptualizes (1) what are unchangeable fragments in the model and (2) the role that various traces in the event log should play when it comes to model repair. A scenario of use is presented that demonstrates the benefits of this conceptualization. The current state of existing process model repair techniques is compared against the proposed concepts. The results show that only two existing techniques partially consider the concepts presented in this paper for model repair.

show abstract

“…The event log should not only be completely replayed to the process model by repair but the accuracy of replay should be improved as much as possible (i.e., if an activity occurs independently or iteratively in the event log, the corresponding activity in the process model should also occur in a consistent manner). Model repair deals with deviation between the event log and process model during replay, which includes two types [12]: i) the observable deviation only occurred in the event log. The repair performance of this deviation involves fitness and precision.…”

Section: A Research Backgroundmentioning

confidence: 99%