Leveraging probabilistic machine learning for subsoil modelling to estimate excavated material volumes

Cotoarbă, Dafydd and Straub, Daniel and Smith, Ian FC; Moreno-Rangel, Alejandro and Kumar, Bimal, eds. (2025) Leveraging probabilistic machine learning for subsoil modelling to estimate excavated material volumes. In: EG-ICE 2025. University of Strathclyde Publishing, GBR. ISBN 9781914241826 (https://doi.org/10.17868/strath.00093291)

Preview

Text. Filename: Cotoarba-etal-EG-ICE-2025-Leveraging-probabilistic-machine-learning-for-subsoil.pdf Final Published Version License: Download (2MB)| Preview

Abstract

The construction of buildings and infrastructure often involves the excavation and transportation of large volumes of soils, which contributes significantly to project costs and environmental impacts. A major challenge in planning the management of excavated materials is the significant uncertainty in quantifying soil volumes before excavation. This uncertainty arises from the limited availability of data, such as borehole soundings and cone penetration tests, as well as the reliance on deterministic modelling approaches. Recent advancements in probabilistic machine learning have enabled the training of models to create probabilistic 3D subsoil models, which explicitly account for uncertainties. In this study we investigate the impact of model selection, such as choice of kernel and hyperparameters, on the resulting 3D geological models. Additionally, we explore how these probabilistic models can improve the management of excavation materials. In a study based on the design of an excavation pit for a metro station in Munich, Germany we show first results on how probabilistic models can inform early decisions regarding machine fleet composition and time estimations.

• Share and Export
  

  ReferData Cite XMLRIOXX2 XMLRDF+N3ASCII CitationEndNoteOpenURL ContextObject in SpanMultiline CSVMODSRefWorksJSONAtomRDF+N-TriplesRIS (EndNote Online, Zotero, Ref manager, etc)EP3 XMLMPEG-21 DIDLRDF+XMLSimple MetadataHTML CitationDublin CoreBibTeXOpenURL ContextObjectMETS
• Item metadata

  Item type:	Book Section
  ID code:	93291
  Dates:	Date Event 1 July 2025 Published 9 June 2025 Accepted
  Subjects:	Fine Arts > Architecture
  Department:	Faculty of Engineering > Architecture
  Depositing user:	Pure Administrator
  Date deposited:	27 Jun 2025 15:08
  Last modified:	01 Jul 2025 11:40
  URI:	https://strathprints.strath.ac.uk/id/eprint/93291