Publications
📚 Publications
STAMM is described in a peer-reviewed paper published in SoftwareX (2026). If STAMM supports your work, please cite it using the reference below. The remaining papers describe soft sensors bundled in the IndPenSim demo registry.
The STAMM paper
STAMM: Soft sensor moniToring and mAintenance framework for Machine learning Models
Carlos Suarez · Alexander Astudillo · Brett Metcalfe · Matthew Crowther · Jasper J. Koehorst · Esteban Castillo · Ariane Bize · David Camilo Corrales
The reference paper for STAMM: an open-source MLOps framework for industrial machine-learning soft sensors. It targets the regime where ground-truth labels arrive offline hours or days late, processes drift slowly and non-stationarily, and models are multi-language. STAMM is validated on an industrial-scale fed-batch penicillin fermentation (IndPenSim) with seven coexisting R and Python soft sensors served through one model registry.
SoftwareX 35 (2026) 102783 · doi:10.1016/j.softx.2026.102783How to cite
If you use STAMM in your research, please cite the SoftwareX paper.
Suarez, C., Astudillo, A., Metcalfe, B., Crowther, M., Koehorst, J. J., Castillo, E., Bize, A., & Corrales, D. C. (2026). STAMM: Soft sensor moniToring and mAintenance framework for Machine learning Models. SoftwareX, 35, 102783. https://doi.org/10.1016/j.softx.2026.102783
@article{suarez2026stamm,
title = {STAMM: Soft sensor moniToring and mAintenance framework for Machine learning Models},
author = {Suarez, Carlos and Astudillo, Alexander and Metcalfe, Brett and
Crowther, Matthew and Koehorst, Jasper J. and Castillo, Esteban and
Bize, Ariane and Corrales, David Camilo},
journal = {SoftwareX},
volume = {35},
pages = {102783},
year = {2026},
issn = {2352-7110},
doi = {10.1016/j.softx.2026.102783},
url = {https://doi.org/10.1016/j.softx.2026.102783}
}
Models in the demo registry
The papers below describe soft sensors included as part of the demo available in the model registry within the IndPenSim project. They don't explicitly mention STAMM, but their models are first-class citizens of the demo.