Home // INTELLI 2016, The Fifth International Conference on Intelligent Systems and Applications // View article

A Method to Build a Production Process Model Prior to a Process Mining Approach

Authors:
Britta Feau
Cédric Schaller
Marion Moliner

Keywords: Manufacturing system; incident detection; process model; event logs; process mining.

Abstract:
While manufacturing plants are becoming more digital, the large amount of data they produce remains under-exploited. In particular, event logs generated by production lines are usually monitored only in case of unusual events, such as production incidents. Process mining, a recent research field at the interface between data mining and process modelling, is specialized in extracting knowledge from event logs about the real process followed by a system. While the process model is usually delivered by the provider of the automaton, specific situations can lead to a lack of global view. This adds extra difficulties to carry out a root cause analysis of the incidents. In this short paper, we propose a simple method to overcome that issue. We built an artificial log from the manual root cause analysis performed after series of production incidents and we then applied process mining techniques to recover an accurate view of the process model.

Pages: 129 to 132

Copyright: Copyright (c) IARIA, 2016

Publication date: November 13, 2016

Published in: conference

ISSN: 2308-4065

ISBN: 978-1-61208-518-0

Location: Barcelona, Spain

Dates: from November 13, 2016 to November 17, 2016