Ikumasa Yoshida, Tokyo City University, iku.yoshida@gmail.com
Shin-ichi Nishimura , Okayama University, theg1786@okayama-u.ac.jp
Yu Otake, Tohoku University, yu.otake.b6@tohoku.ac.jp
Zijun Cao, Wuhan University, zijuncao@whu.edu.cn

For the maintenance and the health monitoring of the geo-structures, many kinds of measurements are conducted. To identify the current states of geo-structures, inverse analyses are required. Also in the data-driven design, or, the data-driven site characterization, the inverse analysis is usefully used.

In the concept of the data assimilation based on the measured data, and the simulation methods, the techniques of the inverse analysis is important. The various filter techniques such as Ensemble Kalman Filter, Unscented Kalman Filter, Particle Filter, and etc, are developed based on the Bayesian concept, and applied for the prediction in the time-space fields. Also, Markov chain Monte Carlo methods are also incorporated into the approaches.

Recently, the concept of Verification and Validation (V&V) is introduced into the design of the structures. In the concept, the relationship between the measured data, and the uncertainty of the simulated results can be identified with the inverse analysis technique. For the design analysis, considering uncertainty effect, simplified analytical models are required. For this task, various surrogate models are proposed. To complete V &V, the combination of the surrogate models, and the inverse analysis methods will be important.

Since the development of the inverse analysis will have more and more demand in the geotechnical engineering, we propose the session to discuss the wide range of the inverse analysis and data assimilation techniques.