Considering Non-sequential Control Flows for Process Prediction with Recurrent Neural Networks

Predictive business process monitoring aims to predict how an ongoing process instance will unfold up to its completion, thereby facilitating proactively responding to anticipated problems. Recurrent Neural Networks (RNNs), a special form of deep learning techniques, gain interest as a prediction technique in BPM. However, non-sequential control flows may make the prediction task more difficult, because RNNs were conceived for learning and predicting sequences of data. Based on an industrial dataset, we provide experimental results comparing different alternatives for considering non-sequential control flows. In particular, we consider cycles and parallelism for business process prediction with RNNs.

[1]  Peter Fettke,et al.  A Multi-stage Deep Learning Approach for Business Process Event Prediction , 2017, 2017 IEEE 19th Conference on Business Informatics (CBI).

[2]  Mike P. Papazoglou,et al.  Formalizing and appling compliance patterns for business process compliance , 2016, Software & Systems Modeling.

[3]  Alessandro Sperduti,et al.  LSTM networks for data-aware remaining time prediction of business process instances , 2017, 2017 IEEE Symposium Series on Computational Intelligence (SSCI).

[4]  Fabrizio Maria Maggi,et al.  Predictive Monitoring of Business Processes , 2013, CAiSE.

[5]  Cesare Pautasso,et al.  Micro-Benchmarking BPMN 2.0 Workflow Management Systems with Workflow Patterns , 2016, CAiSE.

[6]  Klaus Pohl,et al.  Comparing and Combining Predictive Business Process Monitoring Techniques , 2015, IEEE Transactions on Systems, Man, and Cybernetics: Systems.

[7]  Antonello Calabrò,et al.  Monitoring of Business Process Execution Based on Performance Indicators , 2015, 2015 41st Euromicro Conference on Software Engineering and Advanced Applications.

[8]  Yurdaer N. Doganata,et al.  Leveraging path information to generate predictions for parallel business processes , 2015, Knowledge and Information Systems.

[9]  Andreas Metzger,et al.  Risk-Based Proactive Process Adaptation , 2017, ICSOC.

[10]  Sabri Boughorbel,et al.  Optimal classifier for imbalanced data using Matthews Correlation Coefficient metric , 2017, PloS one.

[11]  Andreas Metzger,et al.  Predictive Business Process Monitoring Considering Reliability Estimates , 2017, CAiSE.

[12]  Klaus Pohl,et al.  Accurate Proactive Adaptation of Service-Oriented Systems , 2013, Assurances for Self-Adaptive Systems.

[13]  Marlon Dumas,et al.  Predictive Business Process Monitoring with LSTM Neural Networks , 2016, CAiSE.

[14]  Fabrizio Maria Maggi,et al.  Outcome-Oriented Predictive Process Monitoring , 2017, ACM Trans. Knowl. Discov. Data.

[15]  Antonio Ruiz-Cortés,et al.  Predictive Monitoring of Business Processes: A Survey , 2018, IEEE Transactions on Services Computing.

[16]  Miroslaw Malek,et al.  A survey of online failure prediction methods , 2010, CSUR.

[17]  Jana-Rebecca Rehse,et al.  Predicting process behaviour using deep learning , 2016, Decis. Support Syst..

[18]  Hajo A. Reijers,et al.  Checking Process Compliance on the Basis of Uncertain Event-to-Activity Mappings , 2017, CAiSE.

[19]  J B Brown,et al.  Classifiers and their Metrics Quantified , 2018, Molecular informatics.

[20]  Fabrizio Maria Maggi,et al.  An Eye into the Future: Leveraging A-priori Knowledge in Predictive Business Process Monitoring , 2017, BPM.