STAMM: Soft sensor moniToring and mAintenance framework for Machine learning Models

Abstract

STAMM (Soft sensor moniToring and mAintenance framework for Machine learning Models) is an open source MLOps framework for industrial machine-learning soft sensors. Unlike general-purpose MLOps platforms (MLflow, Kubeflow, Metaflow, ClearML), STAMM targets the regime where ground-truth labels arrive of fline hours or days late, processes exhibit slow non-stationary dynamics, and models are multi-language. The framework comprises five loosely coupled components: a time-series database, workflow orchestrator, language agnostic REST model registry, dashboard with human-in-the-loop labelling, and an extensible drift-detection package. STAMM was validated on an industrial-scale fed-batch penicillin fermentation (IndPenSim) with seven coexisting R and Python soft sensors served through the model registry.

Continuous monitoring of deficit irrigation in avocado across two contrasting rainfall years using sensor networks, telemetry, and machine learning
A. Ferez-Gomez, L. Soler-Escamez, C. Ferrer-Blanco +4
bioRxiv3 Sep 2026PreprintAll rights reservedFull textPlant biologybioRxiv
Enhancing Strawberry Yield Forecasting With Backcasted IoT Sensor Data and Machine Learning
Tewodros Alemu Ayall, Andy Li, Matthew Beddows +2
IEEE Transactions on AgriFood Electronics9 Jul 2026Peer-reviewedLicence not statedFull text5 citedPlant biologyOpenAlex
Smartphone-Assisted Aerogel Colorimetric Sensor Array Coupled with Lightweight Deep Learning for Multi-Meat Freshness Monitoring.
Hanzhi Chen, Xiao Zhou, Caoyuan Zhang +3
Gels (Basel, Switzerland)9 Oct 2026Peer-reviewedCC BYFull textZoology & animal behaviourEurope PMC
Ambient RF energy harvesting using voltage doubler rectifier for battery-free agricultural sensor networks: Design, statistical analysis, and machine learning validation
Md. Atik Hasan Nishat, Prithwiraj Biswas Pallab, Nowrin Jannat +6
PLOS One10 Sep 2026Peer-reviewedCC BY 4.0Full textZoology & animal behaviour StatisticsPLOS
Smart Agriculture Through Internet of Things and Machine Learning: A Survey of Monitoring Techniques, Methods and Applications
Ado Adamou Abba Ari, Emmanuel Baba, Hamayadji Abdoul Aziz +3
IET Wireless Sensor Systems4 Nov 2026Peer-reviewedCC BY-NC-ND1 citedPlant biologyHAL

Cite

BibTeX RIS

BibTeX

@article{suarez2026stamm,
  title = {{STAMM: Soft sensor moniToring and mAintenance framework for Machine learning Models}},
  author = {Carlos Suarez and Alexander Astudillo and Brett Metcalfe and Matthew Crowther and Jasper J Koehorst and Esteban Castillo and Ariane Bize and David Camilo Corrales},
  journal = {SoftwareX},
  year = {2026},
  month = oct,
  doi = {10.1016/j.softx.2026.102783},
  url = {https://hal.inrae.fr/hal-05781695v1},
  note = {CC BY 4.0},
}

RIS

TY  - JOUR
TI  - STAMM: Soft sensor moniToring and mAintenance framework for Machine learning Models
AU  - Carlos Suarez
AU  - Alexander Astudillo
AU  - Brett Metcalfe
AU  - Matthew Crowther
AU  - Jasper J Koehorst
AU  - Esteban Castillo
AU  - Ariane Bize
AU  - David Camilo Corrales
PY  - 2026
DA  - 2026/10/06
JO  - SoftwareX
DO  - 10.1016/j.softx.2026.102783
UR  - https://hal.inrae.fr/hal-05781695v1
AB  - STAMM (Soft sensor moniToring and mAintenance framework for Machine learning Models) is an open source MLOps framework for industrial machine-learning soft sensors. Unlike general-purpose MLOps platforms (MLflow, Kubeflow, Metaflow, ClearML), STAMM targets the regime where ground-truth labels arrive of fline hours or days late, processes exhibit slow non-stationary dynamics, and models are multi-language. The framework comprises five loosely coupled components: a time-series database, workflow orchestrator, language agnostic REST model registry, dashboard with human-in-the-loop labelling, and an extensible drift-detection package. STAMM was validated on an industrial-scale fed-batch penicillin fermentation (IndPenSim) with seven coexisting R and Python soft sensors served through the model registry.
SN  - 2352-7110
ER  -

Abstract

Related

Cite