Comparative study on semi-probabilistic design methods to calibrate load and resistance factors for sliding stability design of caisson breakwaters

• Published in: Ocean engineering 2024-02, Vol.293, p.116573, Article 116573
• Main Authors: Dinh, Huu-Ba, Mac, Van Ha, Doan, Nhu Son
• Format: Article
• Language: English
• Subjects: Caisson sliding stability; Load and resistance factor calibration; LRFD; Optimization-based calibration; Reliability analysis; Semi-probabilistic design
• ISSN: 0029-8018; 1873-5258
• DOI: 10.1016/j.oceaneng.2023.116573

The partial safety factor method commonly necessitates a set of multiplying factors (MFs) to modify the basic variables; hence, when multiple variables are included, it is complicated for design practice. Conversely, the load and resistance factor design (LRFD) format is more practical since it utilizes fewer factors to adjust loads and resistance. This study carries out a comparative study of the two semi-probabilistic design formats and four calibration methods to determine the more suitable design approach and establish associated factors for sliding stability designs of caisson breakwaters under wave and seismic conditions. To fulfill the aforementioned purposes, 12 perforated caisson breakwaters are first selected; thereafter, four calibration methods, which include a closed-form solution and three scenarios based on Monte Carlo simulation (MCS), are executed to determine individual MFs for each case. The final single set of MFs for design practice is calibrated via an optimization process. The present study shows that the two design formats can provide similar solutions. Hence, the LRFD-based concept is recommended to be applied owing to its easier implementation in code calibrations and design practices. Moreover, the performance of the shifting-technique-MCS-based calibration (the simplest simulation) is thoroughly investigated, and it is then found that this approach must be used with care if the target safety considerably varies from the initial safety level. In the present work, an efficient and accurate approximation method to specify the final MFs set is proposed because the optimization-based calibrations to determine the final set of MFs are time-consuming and expensive to evaluate. •Performances of LRFD and PSM were examined and compared.•Similar LRFs were calibrated, although non-unique solutions occurred in DMCS.•Drawbacks of SMCS were pointed out, and SMCS is not suggested in practice.•TDMCS was reported as the most suitable method to calibrate LRFs.•A facile method avoiding optimization problems was proposed to assess final LRFs.