Computer-aided design of formulated products: A bridge design of experiments for ingredient selection
Computer-aided design of formulated products: A bridge design of experiments for ingredient selection
Formulations are ubiquitous in many industries. As formulations are being modified and re-developed to includemore renewable and recyclable ingredients, the speed of formulations development becomes important. Thisstudy expands on the previous work demonstrating successful application of multi-objective Bayesian optimi-zation to design of formulations within a restricted set of the available ingredients. Here we develop an approachthat resolves the un-solved to date problem in algorithmic formulations development, when a subset of in-gredients should be chosen from a larger available pool of suitable ingredients. The new DoE algorithm wasdemonstrated in a workflow making use of a ’make and test’ formulation robots. The developed new DoEprocedure demonstrated an efficient selection of a subset of ingredients from a larger number of the availableones, optimizing their concentration and allowing assignment of differential priorities to the optimizationobjectives.
Design of Experiments, Bayesian optimization, Product design, Gaussian processes, Machine Learning
Cao, Liwei
969e7662-d2ae-4c70-8cb2-11f38e3c68e2
Russo, Danilo
30ac2467-6f46-4842-8dce-0f368e3212dc
Matthews, Emily S
aaab52c6-3e01-44c9-ae35-649e07ab79d5
Lapkin, Alexei
9ef6f6a0-3802-43a4-8425-ffaf03711102
Woods, David
ae21f7e2-29d9-4f55-98a2-639c5e44c79c
7 December 2022
Cao, Liwei
969e7662-d2ae-4c70-8cb2-11f38e3c68e2
Russo, Danilo
30ac2467-6f46-4842-8dce-0f368e3212dc
Matthews, Emily S
aaab52c6-3e01-44c9-ae35-649e07ab79d5
Lapkin, Alexei
9ef6f6a0-3802-43a4-8425-ffaf03711102
Woods, David
ae21f7e2-29d9-4f55-98a2-639c5e44c79c
Cao, Liwei, Russo, Danilo, Matthews, Emily S, Lapkin, Alexei and Woods, David
(2022)
Computer-aided design of formulated products: A bridge design of experiments for ingredient selection.
Computers and Chemical Engineering, 169, [108083].
(doi:10.1016/j.compchemeng.2022.108083).
Abstract
Formulations are ubiquitous in many industries. As formulations are being modified and re-developed to includemore renewable and recyclable ingredients, the speed of formulations development becomes important. Thisstudy expands on the previous work demonstrating successful application of multi-objective Bayesian optimi-zation to design of formulations within a restricted set of the available ingredients. Here we develop an approachthat resolves the un-solved to date problem in algorithmic formulations development, when a subset of in-gredients should be chosen from a larger available pool of suitable ingredients. The new DoE algorithm wasdemonstrated in a workflow making use of a ’make and test’ formulation robots. The developed new DoEprocedure demonstrated an efficient selection of a subset of ingredients from a larger number of the availableones, optimizing their concentration and allowing assignment of differential priorities to the optimizationobjectives.
Text
Paper
- Accepted Manuscript
Text
1-s2.0-S0098135422004161-main
- Version of Record
Text
Supplementary material
- Other
More information
Accepted/In Press date: 25 November 2022
e-pub ahead of print date: 28 November 2022
Published date: 7 December 2022
Keywords:
Design of Experiments, Bayesian optimization, Product design, Gaussian processes, Machine Learning
Identifiers
Local EPrints ID: 473470
URI: http://eprints.soton.ac.uk/id/eprint/473470
ISSN: 0098-1354
PURE UUID: 862e5b21-7358-434e-b620-3532f19eb980
Catalogue record
Date deposited: 19 Jan 2023 17:36
Last modified: 17 Mar 2024 02:51
Export record
Altmetrics
Contributors
Author:
Liwei Cao
Author:
Danilo Russo
Author:
Emily S Matthews
Author:
Alexei Lapkin
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
