Braced Excavation Collapse: Nicoll Highway, Singapore

International Journal of GEOMATE, 2018

This paper demonstrates the use of explicit finite difference program FLAC to numerically simulate staged braced excavation involving various stages of excavation and strutting. The short term undrained condition is considered in the analyses. A FISH script is developed, with automatic mesh generation capacity, to study the behavior of wall deformation and ground settlement during the stages of construction. The numerical model is developed using the case study of Nicoll Highway in Singapore. Both the geotechnical design and the causes of failure of the Nicoll Highway excavation are re-visited through extensive comparisons with existing published data. It is considered as a successful development with the main purpose in mind to develop a numerical model for the analysis of staged excavation in the construction of underground basements. The success of the development is to be used in assisting other design projects in the future.

This paper will describe in details the issues and challenges involved in the procedures for strut omission by observational approach for two case histories from two different projects – Circle Line Contract C824 and Down Town Line Stage 3 Contract C922. The case history of Overrun Tunnel (ORT) of C922 is basically an underground facility building functions as both Railway Facility (Operation Control Centre) and Electrical Substation (ESS) which is to be built next the Expo Station. ORT is located in old alluvium (OA). The proposed underground overrun tunnel is a box structure with dimensions of approximately 23m wide, 25m deep and approximately 440m long. The proposed diaphragm wall function as the earth retaining system (ERSS), it designed for both temporary loading conditions during excavation and permanent load conditions in accordance with LTA Civil Design Criteria. Bottom-up construction sequence is adopted where lateral supports using four (S3 to S6) or six (S1 to S6) layers of steel strutting were installed as excavation progresses downward. The most challenging part is the omission of the last layer of strut S6 for the whole ORT by using observational approach. The case history of C824 Nicoll Highway Station demonstrates that Jet Mechanical Mixing (JMM), if properly installed, has major benefits in controlling the stability and movements induced by deep excavations in soft ground. The reasons can be attributed to the fact that the inner soil column is comprehensively mixed, combined with the attributes of the outer jet grouted column with sufficient overlapping. The whole process undergoes tight quality control and rigorous testing to ensure a continuous and comprehensive slab. In addition to the JMM slab, there is the major benefit of the discrete soil mixing columns formed above the JMM slab during the withdrawal of the auger. This case history also shows that with observational approach, if used appropriately, the design of temporary works can be effectively streamlined to achieve a more economical and yet safe design, This is illustrated by the approach to omit the intermediate layer of strut in the original design after observing the better than expected performance of the JMM. Based on the limited usage to date it is difficult to suggest what parameters should be used for future design. The approach to the back analyses and forward analyses in the observational approach is presented in this paper. This paper will also discuss the design and construction considerations by focusing on the challenge of strut omission by observation approach. The instrumentation monitoring results will also be presented as evaluation of the performance of the ERSS. The site observation and instrumentation result is in line with the forward analysis prediction for the omission of strut. This proposal has helped to expedite the project with a more economical design. With the implementation of observational approach, we will be able to achieve a more sustainable development of underground infrastructure projects.

In the new pneumatic caisson method (NPC), soil excavation and removal is completed remotely by workers on the ground. In 2007, this method was successfully applied in a tunnel shaft in Shanghai. Combined with the construction example, field monitoring and measuring has been conducted. Typical monitored results, such as the working pressure, lateral earth pressure, reaction pressure, and ground movements, were presented and analyzed. In addition, a numerical approach considering the soil disturbance during construction was proposed to predict the soil movements induced by the NPC construction. It was successfully implemented in the three-dimensional finite element method (FEM) codes. Calculated soil movements were examined and verified by the field measurements. In the meantime, these results were compared with the ones obtained from the two-dimensional approach proposed by the authors in the previous study. Results showed that, they agreed well with each other, and in general the three-dimensional analysis results approached the actual situation more closely.

Engineering failures are hard to define. They may form catastrophic disasters that are serious threats to public safety; yet, at other instances, they may be a shortcoming in the behavior or function of structures. A functional shortcoming, unforeseen and unplanned for, of an engineered element, may sometimes be considered a failure while at other times seen as tolerable or even advantageous. Engineers always regard failures as their greatest enemy; ironically, failures may be considered as their foremost teacher as well. Without failures, much of the current advances in design would not have been gained. Studying well-documented failures is important as the findings can be used to improve the designs and performance of structures. In this article, the different aspects related to failures and their prevention are explored. A conceptual model for the cost of failures is also presented. Additionally, some of the well-documented case histories of failures encompassing a wide range of geotechnical practices are highlighted and the learnt lessons are emphasized. It is demonstrated that failures may be substantially reduced by carrying out proper geotechnical site investigations, adopting flexible designs that incorporate observations during construction, applying strict site supervision, implementing geotechnical instrumentation programs and embracing risk management principles.

Journal of Geotechnical and Geoenvironmental Engineering, 2015

The basement of the Stata Center building on the MIT campus required 12.8m deep excavations covering a large open-plan site and underlain by more than 25m of lightly overconsolidated Boston Blue Clay. The excavations were supported by a floating, perimeter diaphragm and braced with a system of internal corner struts, rakers and tieback anchors. The project involved a complex sequence of berms, access ramps and phased construction of the concrete mat foundation. One of the key goals of the design was to limit ground movements in order to prevent damage to adjacent structures including the Alumni Pool building, which was founded on shallow caissons and located less than 1.5m from the south wall. Lateral wall movements and building settlements were closely monitored throughout construction, while photos from a network of webcams located around the open-plan site provide a detailed time history of the construction processes. This paper describes the development of a comprehensive 3D finite element model for the Stata Center basement excavation, which has been enabled by recent advances in available Plaxis 3D software including efficient multicore iterative solving capabilities, importing of geometric data from CAD files, and use of embedded pile elements to represent tieback anchors. The analyses highlight the effects of the 3D excavation and support geometry on wall and ground movements. The base case results using a simple elasto-plastic (MC) soil model with undrained conditions in the clay are generally in very good agreement with measured performance. The effects of refined constitutive modeling and partial drainage within the clay have a secondary role on numerical predictions for this project.

2019

Measurement data from previous excavation works in soft clay deposits are revisited to benchmark a contemporary constitutive soil model, referred to as the Creep-SCLAY1S. The research is motivated by the planned deep excavations for the West Link tunnel in Gothenburg that require analyses of both shortand long-term performance. Unique data on prior excavation works, which were instrumented and monitored during the construction period, will be utilised. In contrast to previous studies, that focused on the construction period, also the long-term performance will be assessed using satellite data. The analyses are carried out using a contemporary rate-dependent model, considering the on-going (background) creep deformations in the clay deposit. The study shows that while the recent model developments result in improved predictions, additional features need to be further developed: most notably, the formulations for the rate-dependent small strain stiffness in loading and unloading stres...

The advance of computer technology has made the finite element method (FEM) more accessible than ever. Many engineers have tried FEM geotechnical software in handling their geotechnical projects. However, like a pilot with inadequate training, it would backfire if he were to fly a sophisticated jet fighter. Engineers with insufficient geotechnical background may gain access to the sophisticated FEM software without realizing the risk behind it. They make mistakes that may lead to the bad performance or even failure of the geotechnical structures. The author himself, along the years of learning and applying the geotechnical FEM software, has made many mistakes. This paper, with Plaxis application as example, tries to elaborate the common mistakes found in applying the FEM geotechnical software in handling excavation problems.

Computers and Geotechnics, 2021

This paper presents a benchmark of a rate-dependent constitutive model for soft soils, implemented in a 2D finite element code, against the response of an instrumented excavation in sensitive clay: Göta Tunnel in Gothenburg. The monitoring data, which comprise time-series of pore water pressures, displacements, earth pressures and strut forces, provide valuable insights of the time-dependent response during the construction period. The longterm response, in terms of the ongoing settlement rates, is assessed using remote sensing data. The results of the numerical simulations demonstrate that the constitutive model, Creep-SCLAY1S, is capable of capturing the observed response. The trends of vertical and horizontal displacements are captured well until the stage of dewatering, and the evolution of pore pressures and earth pressures is computed with high accuracy, excluding peaks in the measurement values arising from pile and rock anchor installation. Most importantly, the results demonstrate that the rate-dependent model enables to model the complete service life of the tunnel, from construction of the excavation to the tunnel operation, with one unified model parameter set. Furthermore, the comparisons highlight the importance of assessing installation effects both in the choice of construction methods and modelling.

This is a thesis on Bangkok MRT works by Dr. Chanaton Surarak as supervised by Prof. A.S.Balasubramaniam

This dissertation is on the geotechnical aspects of the completed Bangkok MRT Blue Line Project and its extension which is currently under design. There were 18 cut and cover subway stations and nearly 22 km of tunnels constructed by the use of earth pressure balanced shield tunnel boring machines. The soil profile model up to depths of 60 to 65 m consists of seven layers: Weathered Crust and Backfill Material; Very Soft to Soft Bangkok Clay; Medium Stiff Clay; Stiff to Hard Clay; Medium Dense to Very Dense Sand; Very Stiff to Hard Clay; and Very Dense Sand. The strength and deformation characteristics of the Bangkok subsoils are determined from laboratory tests (mainly oedometer and triaxial tests) and in-situ field tests (such as vane tests and pressuremeter tests). Additionally, the small strain behaviour is also investigated using Bender element tests in the laboratory and cross hole seismic tests in the field. The soil parameters needed for the deformation analyses are determined for the Mohr Coulomb Model, Soft Soil Model, Hardening Soil Model, and the Hardening Soil Model with Small Strain Stiffness.