Plant-based protein extrusion optimization: Comparison between machine learning and conventional experimental design

High-moisture extrusion (HME) is a promising technique for developing fibrous plant-based meat analogues. In HME, protein-water formulations are passed through a heated twin-screw barrel before solidifying in a cooling die, where complex physicochemical transformations occur, making process optimiza...

Full description

Saved in:
Bibliographic Details
Main Authors: Yingfen Jiang, Noor Irsyad Bin Noor Azlee, Wing Shan Ko, Kaiqi Chen, Bee Gim Lim, Arif Z. Nelson
Format: Article
Language:English
Published: Elsevier 2025-01-01
Series:Current Research in Food Science
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2665927125001881
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849419702250504192
author Yingfen Jiang
Noor Irsyad Bin Noor Azlee
Wing Shan Ko
Kaiqi Chen
Bee Gim Lim
Arif Z. Nelson
author_facet Yingfen Jiang
Noor Irsyad Bin Noor Azlee
Wing Shan Ko
Kaiqi Chen
Bee Gim Lim
Arif Z. Nelson
author_sort Yingfen Jiang
collection DOAJ
description High-moisture extrusion (HME) is a promising technique for developing fibrous plant-based meat analogues. In HME, protein-water formulations are passed through a heated twin-screw barrel before solidifying in a cooling die, where complex physicochemical transformations occur, making process optimization challenging. Traditional approaches like Response Surface Methodology (RSM) require extensive trials and rely on predefined polynomial models, limiting predictive power. In contrast, Bayesian Optimization (BO), a machine learning technique, uses probabilistic surrogate models to efficiently explore parameter spaces and optimize black-box functions with fewer experiments. This study compares RSM and BO for optimizing the mechanical properties of twin-screw extruded meat analogues to replicate chicken breast by varying barrel temperature, water content, and cooling die temperature. To facilitate a direct comparison, BO was constrained to explore within the dataset employed by RSM, although this restriction may limit BO's full optimization potential. Tensile strength was identified as a key property that improved model fitting and predictive accuracy for both RSM and BO. Compared to the 15 experimental trials required by the RSM approach, BO converged on an optimal set of parameters using only 11 of the 15 RSM trials without tensile strength. When tensile strength was included, the output of only 10 trials was needed before convergence was observed. Experimental validation showed BO predictions had lower a prediction error (≤24.5 %) compared to RSM (up to 61.0 %). These findings highlight the potential of superior predictive accuracy and efficiency in optimizing complex pilot-scale food processing systems such as HME through BO.
format Article
id doaj-art-e92d7309c7ed4ce4896d14e0229c017a
institution Kabale University
issn 2665-9271
language English
publishDate 2025-01-01
publisher Elsevier
record_format Article
series Current Research in Food Science
spelling doaj-art-e92d7309c7ed4ce4896d14e0229c017a2025-08-20T03:32:00ZengElsevierCurrent Research in Food Science2665-92712025-01-011110115710.1016/j.crfs.2025.101157Plant-based protein extrusion optimization: Comparison between machine learning and conventional experimental designYingfen Jiang0Noor Irsyad Bin Noor Azlee1Wing Shan Ko2Kaiqi Chen3Bee Gim Lim4Arif Z. Nelson5Food, Chemical and Biotechnology Cluster, Singapore Institute of Technology, 1 Punggol Coast Road, Singapore, 828608, SingaporeFood, Chemical and Biotechnology Cluster, Singapore Institute of Technology, 1 Punggol Coast Road, Singapore, 828608, SingaporeFood, Chemical and Biotechnology Cluster, Singapore Institute of Technology, 1 Punggol Coast Road, Singapore, 828608, SingaporeDepartment of Computer Science, National University of Singapore, 13 Computing Drive, Singapore, 117417, SingaporeFood, Chemical and Biotechnology Cluster, Singapore Institute of Technology, 1 Punggol Coast Road, Singapore, 828608, SingaporeFood, Chemical and Biotechnology Cluster, Singapore Institute of Technology, 1 Punggol Coast Road, Singapore, 828608, Singapore; Corresponding author.High-moisture extrusion (HME) is a promising technique for developing fibrous plant-based meat analogues. In HME, protein-water formulations are passed through a heated twin-screw barrel before solidifying in a cooling die, where complex physicochemical transformations occur, making process optimization challenging. Traditional approaches like Response Surface Methodology (RSM) require extensive trials and rely on predefined polynomial models, limiting predictive power. In contrast, Bayesian Optimization (BO), a machine learning technique, uses probabilistic surrogate models to efficiently explore parameter spaces and optimize black-box functions with fewer experiments. This study compares RSM and BO for optimizing the mechanical properties of twin-screw extruded meat analogues to replicate chicken breast by varying barrel temperature, water content, and cooling die temperature. To facilitate a direct comparison, BO was constrained to explore within the dataset employed by RSM, although this restriction may limit BO's full optimization potential. Tensile strength was identified as a key property that improved model fitting and predictive accuracy for both RSM and BO. Compared to the 15 experimental trials required by the RSM approach, BO converged on an optimal set of parameters using only 11 of the 15 RSM trials without tensile strength. When tensile strength was included, the output of only 10 trials was needed before convergence was observed. Experimental validation showed BO predictions had lower a prediction error (≤24.5 %) compared to RSM (up to 61.0 %). These findings highlight the potential of superior predictive accuracy and efficiency in optimizing complex pilot-scale food processing systems such as HME through BO.http://www.sciencedirect.com/science/article/pii/S2665927125001881Machine learningBayesian optimizationTwin-screw extrusionPlant-based proteinsTensile strengthResponse surface methodology
spellingShingle Yingfen Jiang
Noor Irsyad Bin Noor Azlee
Wing Shan Ko
Kaiqi Chen
Bee Gim Lim
Arif Z. Nelson
Plant-based protein extrusion optimization: Comparison between machine learning and conventional experimental design
Current Research in Food Science
Machine learning
Bayesian optimization
Twin-screw extrusion
Plant-based proteins
Tensile strength
Response surface methodology
title Plant-based protein extrusion optimization: Comparison between machine learning and conventional experimental design
title_full Plant-based protein extrusion optimization: Comparison between machine learning and conventional experimental design
title_fullStr Plant-based protein extrusion optimization: Comparison between machine learning and conventional experimental design
title_full_unstemmed Plant-based protein extrusion optimization: Comparison between machine learning and conventional experimental design
title_short Plant-based protein extrusion optimization: Comparison between machine learning and conventional experimental design
title_sort plant based protein extrusion optimization comparison between machine learning and conventional experimental design
topic Machine learning
Bayesian optimization
Twin-screw extrusion
Plant-based proteins
Tensile strength
Response surface methodology
url http://www.sciencedirect.com/science/article/pii/S2665927125001881
work_keys_str_mv AT yingfenjiang plantbasedproteinextrusionoptimizationcomparisonbetweenmachinelearningandconventionalexperimentaldesign
AT noorirsyadbinnoorazlee plantbasedproteinextrusionoptimizationcomparisonbetweenmachinelearningandconventionalexperimentaldesign
AT wingshanko plantbasedproteinextrusionoptimizationcomparisonbetweenmachinelearningandconventionalexperimentaldesign
AT kaiqichen plantbasedproteinextrusionoptimizationcomparisonbetweenmachinelearningandconventionalexperimentaldesign
AT beegimlim plantbasedproteinextrusionoptimizationcomparisonbetweenmachinelearningandconventionalexperimentaldesign
AT arifznelson plantbasedproteinextrusionoptimizationcomparisonbetweenmachinelearningandconventionalexperimentaldesign