PuSH - Publikationsserver des Helmholtz Zentrums München

Export: TXT Text RIS Endnote (RIS) BIB BibTeX
Roider, J.* ; Nguyen, A.* ; Zanca, D.* ; Eskofier, B.M.

Assessing the performance of remaining time prediction methods for business processes.

IEEE Access 12, 130583-130601 (2024)
Verlagsversion DOI
Open Access Gold
Creative Commons Lizenzvertrag
The prediction of the remaining time for business processes is a major task in predictive process monitoring (PPM). In the last years, various machine learning methods were introduced which reduced error levels steadily. However, the commonly applied metric for optimization and evaluation, the Mean Absolute Error (MAE), has limitations regarding its interpretability. In this work we introduce and evaluate the normalized Mean Absolute Error (nMAE) as an interpretable metric for model evaluation. It accounts for different kinds of label shifts, which are a special type of concept drift that can distort remaining time results. We investigate these concepts in a thorough benchmark study and use them to assess the current state of remaining time prediction for business processes. This includes the evaluation of four different baseline models, identifying the most accurate one. Furthermore, our study compares three different state-of-The-Art methods, namely XGBoost, DA-LSTM, and PGT-Net. In contrary to prior studies we find that there is no significant difference in the performance between these models. Additionally, using the nMAE as evaluation metric we find that these models do not perform reasonably well on a range of event logs. Initial ideas for this behaviour are discussed and consolidated along with other findings from the case study into a comprehensive list motivating future research directions.
Impact Factor
Scopus SNIP
Altmetric
3.400
0.000
Tags
Anmerkungen
Besondere Publikation
Auf Hompepage verbergern

Zusatzinfos bearbeiten
Eigene Tags bearbeiten
Privat
Eigene Anmerkung bearbeiten
Privat
Auf Publikationslisten für
Homepage nicht anzeigen
Als besondere Publikation
markieren
Publikationstyp Artikel: Journalartikel
Dokumenttyp Wissenschaftlicher Artikel
Schlagwörter Business Process ; Graph Neural Network ; Lstm ; Machine Learning ; Predictive Process Monitoring ; Process Mining ; Remaining Time ; Xgboost
Sprache englisch
Veröffentlichungsjahr 2024
HGF-Berichtsjahr 2024
ISSN (print) / ISBN 2169-3536
e-ISSN 2169-3536
Zeitschrift IEEE Access
Quellenangaben Band: 12, Heft: , Seiten: 130583-130601 Artikelnummer: , Supplement: ,
Verlag IEEE
Verlagsort 445 Hoes Lane, Piscataway, Nj 08855-4141 Usa
Begutachtungsstatus Peer reviewed
Institut(e) Institute of AI for Health (AIH)
POF Topic(s) 30205 - Bioengineering and Digital Health
Forschungsfeld(er) Enabling and Novel Technologies
PSP-Element(e) G-540008-001
Förderungen German Research Foundation (DFG)
DOI 10.1109/ACCESS.2024.3459648
WOS ID 001320473100001
Scopus ID 85204216797
Erfassungsdatum 2024-09-27
[➜Korrektur melden]