In traditional factor-of-safety-based or reliability-based geotechnical engineering design, safety requirements are first satisfied, and then a design with lowest cost is usually selected as the final design. However, the lowest cost design may become unsatisfactory if the actual variation of "noise factors" (i.e., "hard-to-control" factors such as soil parameters) greatly exceeds the level of variation assumed during the design phase. The cost optimization becomes irrelevant if the "robustness" of the design against noise factors cannot be assured. A more rational approach is to achieve design robustness which can ultimately reduce the cost. To this end, a "Robust Geotechnical Design" methodology will be developed, which considers robustness explicitly, in addition to the cost and safety requirements. This project will then transform the Robust Design concept into a novel, computer-aided geotechnical engineering design tool. Robust design aims to minimize the effect of uncertainty on the "system" without actually eliminating the sources of uncertainty. The specific objectives of this research are to (1) develop a Robust Geotechnical Design methodology to achieve design safety, robustness, and efficiency, (2) demonstrate the applicability of this new methodology in braced excavations in urban areas, (3) disseminate the new methodology to the engineering profession, and (4) develop human resources by training highly qualified personnel for the geotechnical field. Robust Geotechnical Design is implemented in a fully automated framework through the Markov-Chain Monte Carlo technique. The intellectual innovation of the project includes the development of a new geotechnical design tool that is expected to have a broad positive impact on geotechnical practice. In addition to the novelty of Robust Geotechnical Design, the research addresses an important need in urban construction, namely, improving the safety of braced excavations. Adopting this new design methodology can reduce the effects of "noise factors" and thus, reduce or eliminate the possible adverse social and economic impacts caused by excavation failure and/or excavation-induced damages to the adjacent structures. Open-source computer codes for Robust Geotechnical Design methodology can be a useful learning and development aid to other investigators and engineers.
If successfully executed, this research can refresh geotechnical engineering solutions with the newly developed Robust Geotechnical Design methodology. This research stands to positively impact the design and construction of braced excavations in particular, and geotechnical engineering design practice in general. Improved design methods can benefit society by eliminating or reducing the chance of excavation failure that may cause an adverse social and economic impact. The developed Robust Geotechnical Design methods can potentially be used in related fields beyond geotechnical engineering. Broader impacts are achieved through dissemination of the developed database of case histories, robust design methodology, open-source computer codes, research findings, integration of research and education with special attention to recruiting and nurturing research assistants from underrepresented groups.
