Detecting Sudden and Gradual Drifts in Business Processes from Execution Traces

Maaradji, Abderrahmane; Dumas, Marlon; Rosa, Marcello La; Ostovar, Alireza

doi:10.1109/tkde.2017.2720601

Cited by 81 publications

(101 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…is observation is aligned with previous studies in the eld of process mining, which have shown that concept dri s are common in the control ow of business processes [24,29,30]. Techniques for automated detection and characterization of process control-ow dri s from event logs and event streams are available [24,29,30]. Researchers and practitioners using predictive monitoring methods should consider applying these detection methods, as well as standard statistical tests on the features extracted, to ensure that there is no dri present, which could a ect the performance of the predictive models.…”

Section: Resultssupporting

confidence: 90%

“…Also, although the hyperparameters were optimized using a state-of-the-art hyperparameter optimization technique, it is possible that using more iterations for optimization or a di erent optimization algorithm, other parameter se ings would be found that outperform the se ings used in the current evaluation. Furthermore, the generalizability of the ndings is to some extent limited by the fact that the experiments were performed on a limited number of prediction tasks (24), constructed from nine event logs. Although these are all real-life event logs from di erent application elds that exhibit di erent characteristics, it may be possible that the results would be di erent using other datasets or di erent log preprocessing techniques for the same datasets.…”

Section: Threats To Validitymentioning

confidence: 99%

See 1 more Smart Citation

Outcome-Oriented Predictive Process Monitoring

Teinemaa

Dumas

Rosa

et al. 2019

ACM Trans. Knowl. Discov. Data

Self Cite

263

281

View full text Add to dashboard Cite

Predictive business process monitoring refers to the act of making predictions about the future state of ongoing cases of a business process, based on their incomplete execution traces and logs of historical (completed) traces. Motivated by the increasingly pervasive availability of ne-grained event data about business process executions, the problem of predictive process monitoring has received substantial a ention in the past years. In particular, a considerable number of methods have been put forward to address the problem of outcomeoriented predictive process monitoring, which refers to classifying each ongoing case of a process according to a given set of possible categorical outcomes -e.g., Will the customer complain or not? Will an order be delivered, canceled or withdrawn? Unfortunately, di erent authors have used di erent datasets, experimental se ings, evaluation measures and baselines to assess their proposals, resulting in poor comparability and an unclear picture of the relative merits and applicability of di erent methods. To address this gap, this article presents a systematic review and taxonomy of outcome-oriented predictive process monitoring methods, and a comparative experimental evaluation of eleven representative methods using a benchmark covering 24 predictive process monitoring tasks based on nine real-life event logs.

show abstract

Section: Resultssupporting

confidence: 90%

Section: Threats To Validitymentioning

confidence: 99%

Outcome-Oriented Predictive Process Monitoring

Teinemaa

Dumas

Rosa

et al. 2019

ACM Trans. Knowl. Discov. Data

Self Cite

263

281

View full text Add to dashboard Cite

show abstract

“…A drift is a statistically significant change observed in an event log or in a stream of traces. In [21,22], we introduced a drift detection technique where the basic idea is to perform a statistical test over the distributions of partially-ordered runs observed in two consecutive time windows sliding over a log or a stream of traces. A run represents a set of traces that are equivalent to each other modulo a concurrency relation between event types.…”

Section: Application To Process Drift Detectionmentioning

confidence: 99%

Local Concurrency Detection in Business Process Event Logs

Armas-Cervantes

Dumas

Rosa

et al. 2019

ACM Trans. Internet Technol.

Self Cite

View full text Add to dashboard Cite

Process mining techniques aim at analysing records generated during the execution of a business process in order to provide insights on the actual performance of the process. Detecting concurrency relations between events is a fundamental primitive underpinning a range of process mining techniques. Existing approaches to this problem identify concurrency relations at the level of event types under a global interpretation. If two event types are declared to be concurrent, every occurrence of one event type is deemed to be concurrent to one occurrence of the other. In practice, this interpretation is too coarse-grained and leads to overgeneralization. This paper proposes a finer-grained approach, whereby two event types may be deemed to be in a concurrency relation relative to one state of the process, but not relative to other states. In other words, the detected concurrency relation holds locally, relative to a set of states. Experimental results both with artificial and real-life logs show that the proposed local concurrency detection approach improves the accuracy of existing concurrency detection techniques.

show abstract

“…A significant challenge for adopting concept drift for process mining is to represent behavior in a time-dependent way. The approach reported in [14] uses causal dependencies and tracks them over time windows. The specific challenge is to not only spot a drift but also to classify it.…”

Section: Motivating Examplementioning

confidence: 99%

“…Approaches like ProDrift [14] and Graph Metrics on Process Graphs [22] put an emphasis on requirement R1. The evaluation of ProDrift in [14] shows that two types of drifts are found with high accuracy (sudden and gradual drifts), hence partly addressing requirement R2; note that the authors report high sensitivity of the technique to the choice of the method parameters. The approach relies on the automated detection of changes in business process executions, which are analyzed based on causal dependency relations studied in process mining [24].…”

Section: Requirementsmentioning

confidence: 99%

Balanced multi-perspective checking of process conformance

et al. 2015

View full text Add to dashboard Cite

Organizations maintain process models that describe or prescribe how cases (e.g., orders) are handled. However, reality may not agree with what is modeled. Conformance checking techniques reveal and diagnose differences between the behavior that is modeled and what is observed. Existing conformance checking approaches tend to focus on the control-flow in a process, while abstracting from data dependencies, resource assignments, and time constraints. Even in those situations when other perspectives are considered, the control-flow is aligned first, i.e., priority is given to this perspective. Data dependencies, resource assignments, and time constraints are only considered as "second-class citizens", which may lead to misleading conformance diagnostics. For example, a data attribute may provide strong evidence that the wrong activity was executed. Existing techniques will still diagnose the data-flow as deviatWhen conducting most of the research reported on in this paper, Dr. de Leoni was also affiliated with University of Padua, Italy, and financially supported by the Eurostar-Eureka Project PROMPT (E! 6696). ing, whereas our approach will indeed point out that the control-flow is deviating. In this paper, a novel algorithm is proposed that balances the deviations with respect to all these perspectives based on a customizable cost function. Evaluations using both synthetic and real data sets show that a multi-perspective approach is indeed feasible and may help to circumvent misleading results as generated by classical single-perspective or staged approaches. F. Mannhardt (B)·

show abstract

Detecting Sudden and Gradual Drifts in Business Processes from Execution Traces

Abstract: Business processes are prone to unexpected changes, as process workers may suddenly or gradually start executing a process differently in order to adjust to changes in workload, season or other external factors.

Cited by 81 publications

References 32 publications

Outcome-Oriented Predictive Process Monitoring

Outcome-Oriented Predictive Process Monitoring

Local Concurrency Detection in Business Process Event Logs

Balanced multi-perspective checking of process conformance

Contact Info

Product

Resources

About