Paper contribution

Uncertainty in AI-aided Engineering Geological Modelling

Trevor G. Carter; Giovanna Vessia; Antonio Dematteis; Wayne Barnett; Diego Di Curzio

Proceedings
EUROCK 2026
Pages
1335-1342
Submitted
28 January 2026
Accepted
15 June 2026
Published
12 August 2026

Summary

Abstract

Assessing risk for projects where lack of understanding of geotechnical conditions could have major impact has always been a challenge. Software approaches for formulating electronic engineering geological models (EGMs) are helping improve understanding of the basic geology of complex project sites. However, irrespective of whether the project is for a civil transportation or hydropower tunnel or dam foundation or for an underground or surface mining project, these days assessing geological and geotechnical risk requires detailed evaluation of enormous volumes of data. Sorting through that data to eliminate errors may obscures relevant, small but important facts. The simple model strategy, which is good practice in the geotechnical field, does not apply to digital engineering geological modeling; the representation of a complex geological form (or phenomenon) must capture all the characteristics and details we are interested in. AI’s entry into all electronic platforms is raising other issues and not always reducing uncertainty in the interpretation of geological data. While AI assisted synthesis of data has potential to improve processing, wrongly used it will negatively impact uncertainty. Establishing geological provenance needs proper understanding of domains, in order to allow extrapolation of geological data from known locations into areas with less subsurface information. AI generated options of splined smooth-form geological interpretation surfaces sometimes may be correct, but often can be seriously wrong. Smooth extrapolation may work in some sedimentary environments and perhaps in some zones of extreme metamorphic folding, but for heavily faulted ground such interpretations can be completely erroneous. This paper addresses some of these uncertainty challenges in building engineering geological models, and attempts to quantitatively outlines current limitations. The paper is based on work being undertaken by the C-28 committee of the IAEG aimed at developing methodologies for assessing risk and uncertainty in building electronic EGM’s using modern geological modelling software.

Contributors

Authors

  1. Trevor G. Carter TGC-Geo-Solutions, Canada ORCID 0000-0001-6746-959X
  2. Giovanna Vessia University "G.d'Annunzio" di Chieti-Pescara, Italy
  3. Antonio Dematteis Lombardi Group, Switzerland
  4. Wayne Barnett SRK, Canada
  5. Diego Di Curzio University of Padua, Italy

Index terms

Keywords

Full text

Attachments

Works cited

References

  1. Baynes, F. J. and Parry, S. (2022). Guidelines for the development and application of engineering geological models on projects. International Association for Engineering Geology and the Environment (IAEG) Commission 25 Publication No. 1, 129 pp.
  2. Bond, C. E., Shipton, A. D, Gibbs, A. D. & Jones, S. (2008). Structural models: Optimizing risk analysis by understanding conceptual uncertainty. First Break, 26(6), 65-71. https://doi.org/10.3997/1365-2397.2008006.
  3. Carter T.G. (1992). Prediction and Uncertainties in Geological Engineering and Rock Mass Characterization Assessment. In: Proceedings of the 4th Italian Rock Mechanics Conference, Torino, pp 1.1–1.22.
  4. Carter T.G. (2010) Applicability of classifications for tunnelling—valuable for improving insight, but problematic for contractual support definition or final design. In: Proceedings of the world tunnelling conference (WTC 2010), 36th ITA Congress, Vancouver, Paper00401, Session 6c
  5. Carter T.G., Diederichs M.S., Carvalho J.L. (2008) Application of modified Hoek-Brown transition relationships for assessing strength and post yield behaviour at both ends of the rock competencescale. Proc. 6th Int’l Symp. on Ground Support in Mining and Civil Eng. Constr’n, Cape Town, SAIMM v108., pp 325–338
  6. Carter T.G.& Marinos V (2020). Putting Geological Focus back into Rock Engineering Design. Rock Mech Rock Eng v.53(10): pp.4487–4508 https://doi.org/10.1007/s00603-020-02177-1.
  7. Carter, T.G., & Barnett, W.P. (2021). Improving Reliability of Structural Domaining for Engineering Projects. Rock Mech Rock Eng. 28p. https://doi.org/10.1007/s00603-021-02544-6;ISSN:0723-2632;1434-453X
  8. Dematteis A., Barnett W.P., &Carter T.G., (2023) Self-checking the reliability of engineering geological models. Proc Geo-Risk 2023 conference, Washington, DC. https:// doi. org/ 10. 1061/97807 84484 999
  9. Dematteis, A., Vessia, G., Di Curzio, D., Carter, T.G., Barnett, W.J., Irsch, B., Uzielli, M., &Jack C., (2025). – C28 GdLManaging Engineering Geological Model Uncertainty, International Association of the Engineering Geology and the Environment. (IAEG) Commission 28 Publication No. X, 55pp.
  10. Fookes, P.G., (1997), Geology for Engineers: The Geological Model, Prediction and Performance. Q J Eng Geol 30:293–424.
  11. Grimstad, E. & Barton, N. (1993). Updating of the Q-system for NMT. In Proceedings of the International Symposium on Sprayed Concrete, Fagernes (1993), pp. 46–66.
  12. Hansen T.F., Erharter G.H., Liu Z., &Torresen, J. (2024) A comparative study on machine learning approaches for rock mass classification using drilling data. Applied Computing and Geosciences 24 (2024) 100199.
  13. Hoek, E. (1994). The challenge of input data in rock engineering. ISRM News J. 2(2):23–24.
  14. Hoek, E. (2009). Conclusions Slide 40 of 40 from Keynote presentation on Recent Experiences in Tunnelling. Athens, October. https://slideplayer.com/slide/3470431/12/images/39/CONCLUSIONS.jpg.
  15. Hoek E., Carter T.G., & Diederichs M.S. (2013) Quantification of the geological strength index chart. In: 47th US rock mechanics/geomechanics symposium, San Francisco: ARMA Paper 13-672
  16. Marinos P., & Hoek E. (2000) GSI: a geologically friendly tool for rockmass strength estimation. In: Proceedings of the GeoEng2000 at the international conference on geotechnical and geological engineering, Melbourne, Technomic publishers, Lancaster, Pennsylvania, pp 1422–1446
  17. Palmström A&Broch E (2006) Use and misuse of rock mass classification systems with particular reference to the Q-system.Tunn.Undergr. Space Technol. 21:575–593
  18. Peck, R.B. (1976). Rock foundations for structures. In: Proceedings of the specialty conference on rock engineering for foundations and slopes, vol 2, Boulder, Colorado, pp 1–21.
  19. Salimi, A., Rostami, J., & Moormann, C., (2019) Application of rock mass classification systems for performance estimation of rock TBMs using regression tree and artificial intelligence algorithms. Tunnelling and Underground Space Technology 92 (2019) https://doi.org/10.1016/j.tust.2019.103046
  20. Schlotfeldt P., & Carter T.G., (2018) A new and unified approach to improved scalability and volumetric fracture intensity quantification for GSI and rock mass strength and deformability estimation. Int J Rock Mech Min Sci 110:48–67
  21. Zhao, T., Wang, S., Ouyang, O., Chen, M., Liu, C., Zhang, J., Yu, L., Wang, F., Xie, Y., Li, J., Wang, F., Grunwald, S., Wong, B.M., Zhang, F., Qian, Z., Xu, Y., Yu, C., Han, W., Sun, T., Shao, Z., & Wang L., (2024). Artificial intelligence for geoscience: Progress, challenges, and perspectives, The Innovation, 5(5), 100691. https://doi.org/10.1016/j.xinn.2024.100691.
  22. Zhao G-F., & Wu, Y., (2025). Artificial intelligence in rock mechanics, International Journal of Rock Mechanics & Mining Sciences, 195, 106245. https://doi.org/10.1016/j.ijrmms.2025.106245