Quotation Solti, Andreas, Weske, Mathias. 2015. Prediction of business process durations using non-Markovian stochastic Petri nets. Information Systems (IS) 54, 1-14.


RIS


BibTeX

Abstract

Companies need to efficiently manage their business processes to deliver products and services in time. Therefore, they monitor the progress of individual cases to be able to timely detect undesired deviations and to react accordingly. For example, companies can decide to speed up process execution by raising alerts or by using additional resources, which increases the chance that a certain deadline or service level agreement can be met. Central to such process control is accurate prediction of the remaining time of a case and the estimation of the risk of missing a deadline. To achieve this goal, we use a specific kind of stochastic Petri nets that can capture arbitrary duration distributions. Thereby, we are able to achieve higher prediction accuracy than related approaches. Further, we evaluate the approach in comparison to state of the art approaches and show the potential of exploiting a so far untapped source of information: the elapsed time since the last observed event. Real-world case studies in the financial and logistics domain serve to illustrate and evaluate the approach presented.

Tags

Press 'enter' for creating the tag

Publication's profile

Status of publication Published
Affiliation WU
Type of publication Journal article
Journal Information Systems (IS)
Citation Index SCI
WU Journalrating 2009 A
WU-Journal-Rating new INF-A, STRAT-B, WH-B
Language English
Title Prediction of business process durations using non-Markovian stochastic Petri nets
Volume 54
Year 2015
Page from 1
Page to 14
Reviewed? Y
DOI http://dx.doi.org/10.1016/j.is.2015.04.004

Associations

Projects
Sensor-Enabled Real-World Awareness for Management Information Systems
People
Solti, Andreas (Former researcher)
External
Weske, Mathias (Hasso Plattner Institute, Potsdam, Germany)
Organization
Applied Information Technology with Focus on IT in Organization (Details)
Research areas (Ă–STAT Classification 'Statistik Austria')
5367 Management information systems (Details)
Google Scholar: Search