Applying the observational method to a deep braced excavation using an artificial neural network
Applying the observational method to a deep braced excavation using an artificial neural network
The Observational Method (OM) has long been used in Geotechnical Engineering, and its benefits have been demonstrated in several real cases. However, it has not been widely adopted among practising engineers. There are several reasons for it, some are related to contractual issues and risk sharing between stakeholders, but others are related to technical problems that jeopardise a smooth and successful implementation from an engineering perspective. This paper proposes a framework to implement the OM through machine learning and optimisation algorithms, developing Artificial Neural Network (ANN) models to act as quick-to-evaluate surrogate models coupled with a Genetic Algorithm (GA) to carry out the respective back analyses. The framework is demonstrated through a case study, showing how it can be used to determine soil parameters unambiguously and provides a systematic approach for implementing the OM. The procedure presented enhances the design and construction of geotechnical structures, supporting a more sustainable development within the geotechnical engineering industry.
Jose Ferrero; Agustin Ruiz Lopez; David M. G. Taborda; S. Brasile
10th European Conference on Numerical Methods in Geotechnical Engineering (NUMGE2023)
6. Machine learning and artificial intelligence
https://doi.org/10.53243/NUMGE2023-303
Contact Information
- Secretary General - Dr A M McNamara
- Email: [email protected]
- Phone: +44 20 7040 8154
- City, University of London
- Northampton Square
- London EC1V 0HB. United Kingdom