A multi-fidelity latent modelling framework for predicting wall thickness distributions in stretch blow moulded PET bottles
The International Journal of Advanced Manufacturing Technology · 17.11.2026
DOI: 10.1007/s00170-026-19430-6 ↗Kurz gefasst
Fachartikel in der Zeitschrift The International Journal of Advanced Manufacturing Technology, online erschienen am 17.11.2026. Verfasst von Liam McGovern und 3 weiteren Autor:innen. Die Arbeit nennt 64 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: The Department for the Economy (DfE), NI.
Abstract (Originalsprache)
The production of a bottle through Stretch Blow Moulding depends on several process parameters; both in temperature settings and in blowing process parameters. For a given preform and bottle, the final thickness distributions can vary greatly. Such variations impact the mechanical performance of bottles and ultimately plastic waste. From this study, the relationship between the process parameters and the thickness distribution was investigated. Appropriate dimensionality reduction techniques were applied to efficiently represent both the process parameters and the thickness profiles, forming the inputs and outputs of a regression model. This framework was further enhanced through the incorporation of multiple data sources (virtual and experimental), leading to improved predictive modelling via multi-fidelity regression. Visualisation of the preprocessing steps aided interpretability and provided physical insight into how the process parameters influence the thickness distribution. This research presents a unified workflow, termed PROPEL (PROcess–Property Exploration via Latent modelling), to accurately predict process-property relationships, with potential applications that could advance the production efficiency of the Stretch Blow Moulding process.
Förderung
- The Department for the Economy (DfE), NI
Zitieren
APA
McGovern, L., Nolan, D., Yan, S., & Menary, G. (2026). A multi-fidelity latent modelling framework for predicting wall thickness distributions in stretch blow moulded PET bottles. The International Journal of Advanced Manufacturing Technology. https://doi.org/10.1007/s00170-026-19430-6
Vancouver
McGovern L, Nolan D, Yan S, Menary G. A multi-fidelity latent modelling framework for predicting wall thickness distributions in stretch blow moulded PET bottles. The International Journal of Advanced Manufacturing Technology. 2026. doi:10.1007/s00170-026-19430-6
BibTeX
@article{McGovern2026,
author = {McGovern, Liam and Nolan, Declan and Yan, Shiyong and Menary, Gary},
title = {A multi-fidelity latent modelling framework for predicting wall thickness distributions in stretch blow moulded PET bottles},
journal = {The International Journal of Advanced Manufacturing Technology},
year = {2026},
doi = {10.1007/s00170-026-19430-6}
}
EUVIMED