AI and Geophysics in Modern Geomechanics: Bridging Subsurface Imaging and Design Parameters with Intelligence
Year: 2026
Type: book-chapter
Abstract: Geotechnical engineering has long confronted a spatial paradox at its heart: large-scale infrastructure is routinely designed on the basis of sparse, point-scale subsurface measurements. This chapter examines how convergence of geophysical methods with artificial intelligence (AI) helps resolve that paradox by delivering continuous, engineering-scale subsurface intelligence. We address the Representative Elementary Volume (REV) problem, the conversion of dynamic to static moduli, and the estimation of geomechanical properties across multiple scales. A systematic treatment is provided of how geophysical outputs, including shear wave velocity (Vs), compressional wave velocity (Vp), and electrical resistivity (ERT), are translated into actionable engineering design parameters such as bearing capacity, settlement, slope stability coefficients, and foundation stiffness. Two case studies illustrate the workflow. Note : Case Study 1 (AI-enhanced MASW for foundation design) and Case Study 2 (ERT with CNN for karst void detection) are presented as engineering scenarios developed from field conditions and published performance benchmarks. They are not verbatim accounts of projects. Cost figures, detection rates, and dimensional data indicate typical outcomes and should not be cited as site-specific verified results without reference to the sources listed in the References. The scenarios are structured to let readers follow the decision-making workflow while protecting project confidentiality. Where numerical claims draw on published studies, references are provided in the text. The chapter also examines critical limitations, including data non-uniqueness, model bias, and the risks of uncritical reliance on automated interpretation, before concluding with frontier technologies such as Physics-Informed Neural Networks (PINNs) and Distributed Acoustic Sensing (DAS).
Source: IntechOpen eBooks
Authors: Yasir Yousif, Abdelmahmoud Elsamani
Language: English
Subfield: Safety, Risk, Reliability and Quality
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7167674826