Bright Spark Lecture
Uncertainty quantification in data-driven geotechnical stratigraphic modeling
Assistant Professor, Department of Civil and Environmental Engineering, University of Dayton
Abstract
Industry 4.0 will bring about a new Geotech digital era dominated by data science and sensing technologies. In alignment with this trend, data-driven geotechnics is an emerging research field that contributes to the digital transformation of geotechnical engineering. In order to practice precision design and construction toward better sustainability and resiliency, geotechnical reliability and risk analysis are sound only when the source uncertainty – the interpretation of observed data–is well quantified. Yet at present, it still heavily depends on engineers' subjective experience and may result in a less- or over-conservative design/decision. The challenges are two-fold: 1) how to better interpret geotechnical data that are multivariate, sparse, incomplete, and potentially corrupted in an algorithmic and smart manner at probed points and 2) conditional on observed data, how to infer and model geotechnical information at vast unobserved locations accurately with quantified uncertainty. In this study, the above two challenges are addressed to a certain extent using a novel Bayesian approach for 2D soil stratigraphic interpretation. This new approach builds upon the author's previously developed one-dimensional hidden Markov random field (HMRF) model and 2D anisotropic Markov random field (MRF) simulation algorithm. Bayesian machine learning is adopted to jointly perform parameter estimation and stochastic simulation of soil stratigraphic profiles. We have tested two different types of site investigation data: borehole logs and cone penetration testing soundings. The advantages of the developed approach can be summarized into three points: 1) inferring stratigraphic profile and associated uncertainty in an automatic and fully unsupervised manner; 2) relatively lower computational cost; 3) both aleatoric and epistemic uncertainties are taken into consideration. This paper contributes to the techniques leveraging valuable data assets for informed decision-making in geotechnical engineering risk management.
Biography
Dr. Hui (Jack) Wang joined the University of Dayton in 2018 as an assistant professor with the Department of Civil and Environmental Engineering. He obtained his BS and MS degrees in civil engineering from Tongji University and received his Ph.D. degree at the University of Akron in geotechnical engineering. Before his faculty appointment, Dr. Wang has three years of research experience in machine learning and computational geosciences at the RWTH Aachen University in Germany. His research focuses on the opportunities in the multidisciplinary fields spanning machine learning, geotechnical/geological subsurface modeling, smart infrastructure, and reliability & risk assessment. He is a member of ISSMGE TC304 (Engineering Practice of Risk Assessment and Management), and ASCE\Geo-Institute Technical committee: Risk Assessment and Management. He is a reviewer for all major international journals of geotechnical engineering, civil and infrastructure engineering, and engineering geology. He is also on the editorial board of the Journal of Pipeline Science and Engineering.
Assistant Professor, Department of Civil and Environmental Engineering, University of Dayton
Abstract
Industry 4.0 will bring about a new Geotech digital era dominated by data science and sensing technologies. In alignment with this trend, data-driven geotechnics is an emerging research field that contributes to the digital transformation of geotechnical engineering. In order to practice precision design and construction toward better sustainability and resiliency, geotechnical reliability and risk analysis are sound only when the source uncertainty – the interpretation of observed data–is well quantified. Yet at present, it still heavily depends on engineers' subjective experience and may result in a less- or over-conservative design/decision. The challenges are two-fold: 1) how to better interpret geotechnical data that are multivariate, sparse, incomplete, and potentially corrupted in an algorithmic and smart manner at probed points and 2) conditional on observed data, how to infer and model geotechnical information at vast unobserved locations accurately with quantified uncertainty. In this study, the above two challenges are addressed to a certain extent using a novel Bayesian approach for 2D soil stratigraphic interpretation. This new approach builds upon the author's previously developed one-dimensional hidden Markov random field (HMRF) model and 2D anisotropic Markov random field (MRF) simulation algorithm. Bayesian machine learning is adopted to jointly perform parameter estimation and stochastic simulation of soil stratigraphic profiles. We have tested two different types of site investigation data: borehole logs and cone penetration testing soundings. The advantages of the developed approach can be summarized into three points: 1) inferring stratigraphic profile and associated uncertainty in an automatic and fully unsupervised manner; 2) relatively lower computational cost; 3) both aleatoric and epistemic uncertainties are taken into consideration. This paper contributes to the techniques leveraging valuable data assets for informed decision-making in geotechnical engineering risk management.
Biography
Dr. Hui (Jack) Wang joined the University of Dayton in 2018 as an assistant professor with the Department of Civil and Environmental Engineering. He obtained his BS and MS degrees in civil engineering from Tongji University and received his Ph.D. degree at the University of Akron in geotechnical engineering. Before his faculty appointment, Dr. Wang has three years of research experience in machine learning and computational geosciences at the RWTH Aachen University in Germany. His research focuses on the opportunities in the multidisciplinary fields spanning machine learning, geotechnical/geological subsurface modeling, smart infrastructure, and reliability & risk assessment. He is a member of ISSMGE TC304 (Engineering Practice of Risk Assessment and Management), and ASCE\Geo-Institute Technical committee: Risk Assessment and Management. He is a reviewer for all major international journals of geotechnical engineering, civil and infrastructure engineering, and engineering geology. He is also on the editorial board of the Journal of Pipeline Science and Engineering.
