On the use of surrogate modelling in parameter selection

Ronald B.J. Brinkgreve; Majd Ahmad

On the use of surrogate modelling in parameter selection

Ronald B.J. Brinkgreve^*
, Majd Ahmad

^*Corresponding author for this work

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

3 Downloads (Pure)

Abstract

Surrogate modelling is an application of Artificial Intelligence in which a Machine Learning model is trained to produce results that are originally obtained from a more complicated (numerical) model, especially when numerous such results are required. This article demonstrates two examples of using surrogate modelling in geotechnical applications involving (constitutive) model parameter selection as part of model fitting (parameter optimization) and probabilistic analysis. It gives some practical ideas on how to maintain the consistency of parameter sets in the case of parameter inter-dependencies, how to improve the quality of the surrogate in the case of large parameter variations, and how to improve model fitting to a target curve.

Original language	English
Title of host publication	Conference Proceedings Workshop Numerical Methods in Geotechnics 2026
Subtitle of host publication	AI in geotechnics, deformaton based design, numerical optimization
Editors	J. Grabe
Publisher	Hamburg University of Technology
Pages	113-125
Number of pages	13
ISBN (Print)	978-3-936310-66-5
Publication status	Published - 2026
Event	Workshop numerical methods in geotechnics 2026: AI in geotechnics, deformation based design, numerical optimization - Hamburg University of Technology - Schwarzenberg-Campus Building H, Hamburg, Germany Duration: 17 Sept 2026 → 18 Sept 2026 https://tore.tuhh.de/entities/event/3c40ba2d-5756-4115-abf5-a6c5498a0d4d

Conference

Conference	Workshop numerical methods in geotechnics 2026
Abbreviated title	NiMG 2026
Country/Territory	Germany
City	Hamburg
Period	17/09/26 → 18/09/26
Internet address	https://tore.tuhh.de/entities/event/3c40ba2d-5756-4115-abf5-a6c5498a0d4d

Access to Document

On the use of surrogate modelling in parameter selectionFinal published version, 752 KBLicence: CC BY

Cite this

@inproceedings{151fa9b311be46c3aca2ba5e3e5510e0,

title = "On the use of surrogate modelling in parameter selection",

abstract = "Surrogate modelling is an application of Artificial Intelligence in which a Machine Learning model is trained to produce results that are originally obtained from a more complicated (numerical) model, especially when numerous such results are required. This article demonstrates two examples of using surrogate modelling in geotechnical applications involving (constitutive) model parameter selection as part of model fitting (parameter optimization) and probabilistic analysis. It gives some practical ideas on how to maintain the consistency of parameter sets in the case of parameter inter-dependencies, how to improve the quality of the surrogate in the case of large parameter variations, and how to improve model fitting to a target curve.",

author = "Brinkgreve, \{Ronald B.J.\} and Majd Ahmad",

year = "2026",

language = "English",

isbn = "978-3-936310-66-5",

pages = "113--125",

editor = "J. Grabe",

booktitle = "Conference Proceedings Workshop Numerical Methods in Geotechnics 2026",

publisher = "Hamburg University of Technology",

note = "Workshop numerical methods in geotechnics 2026 : AI in geotechnics, deformation based design, numerical optimization, NiMG 2026 ; Conference date: 17-09-2026 Through 18-09-2026",

url = "https://tore.tuhh.de/entities/event/3c40ba2d-5756-4115-abf5-a6c5498a0d4d",

}

Brinkgreve, RBJ & Ahmad, M 2026, On the use of surrogate modelling in parameter selection. in J Grabe (ed.), Conference Proceedings Workshop Numerical Methods in Geotechnics 2026: AI in geotechnics, deformaton based design, numerical optimization. Hamburg University of Technology, pp. 113-125, Workshop numerical methods in geotechnics 2026, Hamburg, Hamburg, Germany, 17/09/26.

On the use of surrogate modelling in parameter selection. / Brinkgreve, Ronald B.J.; Ahmad, Majd.
Conference Proceedings Workshop Numerical Methods in Geotechnics 2026: AI in geotechnics, deformaton based design, numerical optimization. ed. / J. Grabe. Hamburg University of Technology, 2026. p. 113-125.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - On the use of surrogate modelling in parameter selection

AU - Brinkgreve, Ronald B.J.

AU - Ahmad, Majd

PY - 2026

Y1 - 2026

N2 - Surrogate modelling is an application of Artificial Intelligence in which a Machine Learning model is trained to produce results that are originally obtained from a more complicated (numerical) model, especially when numerous such results are required. This article demonstrates two examples of using surrogate modelling in geotechnical applications involving (constitutive) model parameter selection as part of model fitting (parameter optimization) and probabilistic analysis. It gives some practical ideas on how to maintain the consistency of parameter sets in the case of parameter inter-dependencies, how to improve the quality of the surrogate in the case of large parameter variations, and how to improve model fitting to a target curve.

AB - Surrogate modelling is an application of Artificial Intelligence in which a Machine Learning model is trained to produce results that are originally obtained from a more complicated (numerical) model, especially when numerous such results are required. This article demonstrates two examples of using surrogate modelling in geotechnical applications involving (constitutive) model parameter selection as part of model fitting (parameter optimization) and probabilistic analysis. It gives some practical ideas on how to maintain the consistency of parameter sets in the case of parameter inter-dependencies, how to improve the quality of the surrogate in the case of large parameter variations, and how to improve model fitting to a target curve.

UR - https://doi.org/10.15480/882.18143

M3 - Conference contribution

SN - 978-3-936310-66-5

SP - 113

EP - 125

BT - Conference Proceedings Workshop Numerical Methods in Geotechnics 2026

A2 - Grabe, J.

PB - Hamburg University of Technology

T2 - Workshop numerical methods in geotechnics 2026

Y2 - 17 September 2026 through 18 September 2026

ER -

On the use of surrogate modelling in parameter selection

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this