Predictive business process monitoring with LSTM neural networks

N. Tax, I. Verenich, M. La Rosa, M. Dumas

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

242 Citations (SciVal)
5 Downloads (Pure)


Predictive business process monitoring methods exploit logs of completed cases of a process in order to make predictions about running cases thereof. Existing methods in this space are tailor-made for specific prediction tasks. Moreover, their relative accuracy is highly sensitive to the dataset at hand, thus requiring users to engage in trial-and-error and tuning when applying them in a specific setting. This paper investigates Long Short-Term Memory (LSTM) neural networks as an approach to build consistently accurate models for a wide range of predictive process monitoring tasks. First, we show that LSTMs outperform existing techniques to predict the next event of a running case and its timestamp. Next, we show how to use models for predicting the next task in order to predict the full continuation of a running case. Finally, we apply the same approach to predict the remaining time, and show that this approach outperforms existing tailor-made methods.
Original languageEnglish
Title of host publicationAdvanced Information Systems Engineering : 29th International Conference, CAiSE 2017, Essen Germany, June 12-16, 2017. Proceedings
EditorsKlaus Pohl, Eric Dubois
Number of pages16
ISBN (Electronic)978-3-319-59536-8
ISBN (Print)978-3-319-59535-1
Publication statusPublished - 2017
Event29th International Conference on Advanced Information Systems Engineering (CAiSE 2017) - Essen, Germany
Duration: 12 Jun 201716 Jun 2017
Conference number: 29

Publication series

ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference29th International Conference on Advanced Information Systems Engineering (CAiSE 2017)
Abbreviated titleCAiSE'17
Internet address


Dive into the research topics of 'Predictive business process monitoring with LSTM neural networks'. Together they form a unique fingerprint.

Cite this