Submitted by Angela Walters on Mon, 29/10/2018 - 13:09
October Monthly Paper - 'A Bayesian Definition of 'Most Probable' Parameters' by Yingyan Jin, Dr Giovanni Biscontin and Prof. Paolo Gardoni
This month's paper from CDBB is a Journal Paper by Yingyan Jin, Giovanna Biscontin, and Paolo Gardoni. It is based on the work carried out as part of the CDBB-funded mini-projects, on Adaptive Design of Supported Excavations.
The abstract for the paper is below:
Since guidelines for choosing ‘most probable' parameters in ground engineering design codes are vague, concerns are raised around their definition, as well as the associated uncertainties. This paper introduces Bayesian inference for a new rigorous approach to obtain the estimates of the most probable parameters based on observations collected during construction. Following the review of optimization-based methods that can be used in back analysis, such as gradient descent and neural networks, a probabilistic model is developed using Clough and O’Rourke’s (1980) method for retaining wall design. Sequential Bayesian inference is applied to a staged excavation project to examine the applicability of the proposed approach and illustrate the process of back analysis.
Full paper: CDBB_JP_004
DOI reference: https://www.repository.cam.ac.uk/handle/1810/280616
Contact corresponding author: Yingyan Jin - yj270@cam.ac.uk