Towards a machine learning operations (MLOps) soft sensor for real-time predictions in industrial-scale fed-batch fermentation

Aug 5, 2026

Upload by Asher

Real-time predictions in fermentation processes enable continuous monitoring and control, yet online measurements are often constrained by sensing availability and feasibility. Soft sensors translate available measurements into key variables such as product yield and quality, and machine learning variants can exploit rich historical data. This work presents a proof concept MLOps approach for end-to-end soft sensor lifecycle automation in industrial fed-batch fermentation. Using the IndPenSim dataset (100 batches), an LSTM soft sensor predicts penicillin concentration, with batches 91–100 evaluating concept drift and triggering retraining via PSI-based alerting.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages14

Reading Time35 min

AccessFree to download

Introduction

Industrial biotechnology

Soft sensors and machine learning for fermentation

MLOps for end-to-end lifecycle management

VU Research Portal Towards a machine learning operations (MLOps) soft sensor for real-time predictions in industrial-scale fed-batch fermentation Metcalfe, Brett; Acosta-Pavas, Juan camilo; Robles-Rodriguez, Carlos eduardo; Georgakilas, George k. ; Dalamagas, Theodore; Aceves-Lara, Cesar arturo; Daboussi, Fayza; Koehorst, Jasper j; Corrales, David camilo published in Computers & Chemical Engineering 2025 DOI (link to publisher) 10.1016/j.compchemeng.2024.108991 document version Publisher's PDF, also known as Version of record document license Article 25fa Dutch Copyright Act Link to publication in VU Research Portal citation for published version (APA) Metcalfe, B. , Acosta-Pavas, J. C. , Robles-Rodriguez, C. E. , Georgakilas, G. K. , Dalamagas, T. , Aceves-Lara, C. A. , Daboussi, F. , Koehorst, J. J. , & Corrales, D. C. (2025) . Towards a machine learning operations (MLOps) soft sensor for real-time predictions in industrial-scale fed-batch fermentation. Computers & Chemical Engineering, 194, 1-13 . Article 108991. [https://doi.org/10.1016/j.compchemeng.2024.108991](https://doi.org/10.1016/j.compchemeng.2024.108991) General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. E-mail address: [vuresearchportal.ub@vu.nl](vuresearchportal.ub@vu.nl) [Download date: 01](Download date: 01) . Oct. 2025 Computers and Chemical Engineering 194 (2025) 108991 Contents lists available at ScienceDirect Computers and Chemical Engineering journal [homepage:](homepage: www.elsevier.com/locate/compchemeng)[ www.elsevier.com/locate/compchemeng](homepage: www.elsevier.com/locate/compchemeng) | Towards a machine learning operations (MLOps) soft sensor for real-time predictions in industrial-scale fed-batch fermentation | | | | | --- | --- | --- | --- | | Brett Metcalfe a,b , Juan Camilo Acosta-Pavas c , Carlos Eduardo Robles-Rodriguez c ,
George K. Georgakilasd, Theodore Dalamagas d , Cesar Arturo Aceves-Lara c, Fayza Daboussi e,
Jasper J Koehorst a,f, David Camilo Corrales c,*
a Laboratory of Systems and Synthetic Biology, Wageningen University & Research, Wageningen, The Netherlands b Department of Earth Sciences, Faculty of Science, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands c TBI, Universit´e de Toulouse, CNRS, INRAE, INSA, Toulouse, France
d Information Management Systems Institute (IMSI), ATHENA Research Center, Athens 15125, Greece e INRAE, UMS (1337) TWB, 135 Avenue de Rangueil, Toulouse 31077, France
f UNLOCK Large Scale Infrastructure for Microbial Communities, Wageningen University & Research, Gelderland, Wageningen, The Netherlands | | | | | A R T I C L E I N F O | | A B S T R A C T | | | Keywords:
Machine learning pipelines MLOps
LTSM
Soft-sensors Concept drift IndPenSim | | Real-time predictions in fermentation processes are crucial because they enable continuous monitoring and control of bioprocessing. However, the availability of online measurements is limited by the availability and feasibility of sensing technology. Soft sensors - or software sensors that convert available measurements into measurements of interest (product yield, quality, etc.) - have the potential to improve efficiency and product quali

FAQs

Why are real-time predictions important in industrial-scale fed-batch fermentation?

They support continuous monitoring and control of bioprocessing, improving operational responsiveness. Limited sensing can restrict direct access to key variables, motivating predictive approaches.

What is the proposed role of MLOps in the soft sensor lifecycle?

MLOps automates the end-to-end soft sensor workflow from development and deployment to maintenance and monitoring. It ensures operational readiness beyond offline model evaluation.

How is concept drift detected and handled in the study?

The LSTM soft sensor is assessed using batches containing process deviations (91–100). Performance degradation below a threshold is detected using the Population Stability Index (PSI), which triggers retraining via the retraining pipeline.

Aug 5, 2026

Upload by Asher