Adaptive management techniques provide a means to incorporate advances in sensor development, information technology, and numerical analyses to a variety of problems in geotechnical engineering. If one wants to predict and subsequently evaluate the overall performance of a design, the ?observational method? espoused by Peck is a well-known framework wherein construction and design procedures and details are adjusted based upon observations and measurements made as construction proceeds. Adaptive management techniques allow one to automate the observational approach so that quantitative information can be distributed to interested shareholders in a timely enough fashion to be of use in a number of applications. This methodology has particular application when extending civil infrastructure into the subsurface environment. The increase in population density, as well as new imperatives for infrastructure investment, homeland security, energy conservation and sustainable design, will require major advances in our ability to create underground space more efficiently and safely.
Many factors affect the ground movements caused by excavations, including stratigraphy, soil properties, support system details, construction activities, contractual arrangements and workmanship. While numerical simulations have become more common when analyzing ground response to excavations as part of the design process, finite element predictions contain uncertainties related to soil properties, support system details, and construction procedures. These factors are explicitly considered when applying adaptive management methods to the problem, although the methodology has its limitations as subsequently discussed.
Stricter limits on allowable ground movements associated with deep excavations are being imposed in many locales by either regulatory agencies or by recognition of adverse effects of excessive ground movements. Excavations in some urban areas are being subjected to movement limitations that are much smaller than even just 5 years ago. For example, excavation-induced ground movements in the Seattle area are limited to 1 inch for cuts at deep as 70 ft. Requirements for excavations in Chicago are now targeted for maximum ground movements of 1-1/2 to 2 inches, down from 4 inches just over a decade ago. These issues are compounded in Chicago by the fact that excavations now are being made to depths of 75 ft, much deeper than the typical depth of 40 ft. Many excavations in the Boston area are limited to 1 inch of ground movement. Given that the subsurface conditions at Boston and Chicago consist of relatively soft clays, these requirements present a challenge in excavation support design and construction. Consequently, the state-of-the-art of predicting ground deformations has reached a point where major advances in practice are required to make accurate design assessments when movements are limited to such small amounts. These advances also are needed to make the adaptive management approach applicable to these types of problems.
The purpose of this research is to extend the adaptive management approach so that it applies to a range of problems where ground deformations must be limited to prevent damage to adjacent buildings and other infrastructure. In particular it is proposed to quantify the relative effects of small strain non-linearity of soils, non-linear stiffness of walls and shrinkage of floor slabs used in top-down construction on the deformations associated with excavations. This research will include laboratory experiments on block samples cut from excavations in Chicago and Boston to characterize the constitutive behavior of the soils with emphasis on the small strain responses, detailed field experiments at deep excavations where ground deformation and structural responses of the support system are measured and related to the construction activities at the site, and finite element simulations and after-the-fact adaptive management evaluations using inverse analysis based on observed field observations.
Adaptive management techniques provide a means to incorporate advances in sensor development, information technology, and numerical analyses and apply it to a variety of problems in geotechnical engineering. If one wants to predict and subsequently evaluate the overall performance of a design, a procedure that incorporates an evaluation of the results of the predictive analysis must be defined. This procedure is usually referred to as the "observational method," a framework wherein construction and design procedures and details are adjusted based upon observations and measurements made as construction proceeds. While the observational method is conceptually very helpful, it is quite difficult to use observed movements for controlling construction or to judge quantitatively how well the work is proceeding in a timely enough fashion to be of use in many applications. Northwestern University researchers have applied adaptive management techniques to design and construction of supported excavations, a problem particularly suited to its use. Many factors affect the ground movements caused by excavations, including stratigraphy, soil properties, support system details, construction activities, contractual arrangements and workmanship. While numerical simulations have become more common in engineering practice to analyze ground response to excavations as part of the design process, finite element predictions contain uncertainties related to soil properties, support system details, and construction procedures. These factors can be considered explicitly when applying adaptive management methods to the problem. Stricter limits on allowable ground movements associated with deep excavations are being imposed in many locales by either regulatory agencies or by recognition of adverse effects of excessive ground movements. Given that the subsurface conditions in many urban centers consist of relatively soft clays, these requirements present a challenge in design and construction of excavation support systems. The state of the art of predicting ground deformations has reached a point where major advances in practice are required to reliably make accurate design assessments when movements are limited small amounts. The main two sources of uncertainties in the analysis are the structural evaluation of the affected building and the movement prediction. The latter depends on accurate representation of the stress-strain characteristics of the soil, and given the uncertainty associated with its definition, we have developed and refined adaptive management techniques to address this issue. In particular, we quantified the relative effects of small strain non-linearity of soils, non-linear stiffness of walls and shrinkage of floor slabs used in top-down construction on the deformations associated with excavations. This research included extensive laboratory experiments on block samples cut from excavations in Chicago to characterize the constitutive behavior of the soils with emphasis on the small strain responses, detailed field experiments at three deep excavations where ground deformation and structural responses of the support system are measured and related to the construction activities at the site, and finite element simulations and after-the-fact adaptive management evaluations using inverse analysis based on observed field observations. The partnerships between Northwestern University and their industrial contributors, AECOM at the OMPW project, STS Consultants at the Block 37 and Hayward Baker, Inc. at the Jones School project, allowed the researchers to relate conventional date collected during site investigations, research quality experimental results and state-of-the-art numerical simulation to full-scale performance of several geo-structures to advance to the state-of-the-art and practice of geotechnical engineering. Successful completion of this research improved the state-of-the-art and practice of predicting and controlling ground movements associated with supported excavations and tunneling operations. Analyses of the results of the observations at the three sites improved adaptive management design procedures to allow a designer to minimize the effects of deformations on adjacent structures and utilities. The results of the experimental program conducted on block samples cut from the Jones and Block 37 excavations provided detailed stress-strain data from very small through failure strains. This data set allowed us to develop a method to find parameters for a sophisticated constitutive model based on data commonly collected during site investigation, thus making this model accessible to engineers for use in practice. Results of performance data from the Block 37 and especially the OMPW project, allowed us to define the interrelation between concrete slab shrinkage, consolidation and ground movements in top down excavations. Results of performance monitoring and analyses of the excavation for the William Jones High School allowed us to identify the conditions whereupon minor cracks were initiated on the buildings.
