The suppression of innovation and matters of regulations and litigation too often stem from our traditional approach to the engineering design process, in which problems are considered and decisions made during design. Decisions formed from inadequate data and inadequate evaluation of risk become fixed in concrete, steel, and earth. There is a reluctance to change design decisions once made; a reluctance that increases once construction is under way, and becomes a resistance once a facility is in operation. Yet for many projects, decisions to modify a design need to be made during construction as well as during operation, particularly when unexpected behavior occurs. Data from long‐term monitoring should be integrated into the design process provide a basis for future decisions and maintain a facility in a functional state consistent with its intended purpose. A planned approach to decision making over time that draws on long‐term field measurements for input, with planned analysis of the measurements and appropriate contingent actions, is sought. A monitored‐decision process provides a means to gain knowledge, be innovative, and mitigate adverse relationships between parties involved in the ownership, construction, and operation of a facility.

The installation of the pressuremeter probe in the ground causes disturbance to the adjacent soil, resulting in the formation of an annular zone of remolded soil around the probe. In the conventional pressuremeter test, disturbance is caused by temporary unloading of the borehole until the probe is inserted and inflated. In this study the disturbance was modelled by (1) Assuming a remolded annulus of soil around the probe; and (2) considering initial unloading and reloading of the cavity wall. A parametric study was conducted by varying the thickness of the remolded zone and the amount of initial unloading. The results indicate that the error in the undrained strength due to initial unloading can be higher than that due to the remolded zone. The modulus is affected by both the unloading and the size of the disturbed annulus. Further, an examination of field data indicates that strain‐softening soils are affected most by the disturbance caused by initial unloading of the cavity wall.

Soil nailing is an in situ soil reinforcement technique that has been used during the last two decades mainly in France and Germany to retain excavations or stabilize slopes. Design of soil‐nailed systems has been traditionally done using slope‐stability analysis methods. These methods have been developed to incorporate the effect of the available tension and shear resistance of the passive reinforcements on the slope stability. However, they provide only a global safety factor. This paper presents a kinematical limit analysis design approach that provides a rational estimate of maximum tension and shear forces mobilized in each reinforcement. To verify the applicability of the method, the predicted forces are compared with those measured in both laboratory models and full‐scale structures. The proposed design approach is also used to analyze the various failure mechanisms observed on model walls and predicted critical model heights are compared with experimental results.

The results of loading tests are used to develop empirical correlations between the point‐bearing resistance of expanded‐base piles and the results of standard penetration tests. The resistance is shown to be proportional to depth for piles having shaft lengths not exceeding 10 base diameters. The results also show that there is no significant difference between the point resistance of piles in relatively homogeneous soils and piles founded in a bearing stratum overlain by soft, compressible soil at the same overall depth/base diameter ratio measured from ground surface. Values of the bearing capacity factor $N_q$ were calculated by dividing the point resistance at failure by the effective overburden pressure at the level of the maximum cross section of the base. $N_q$ was found to decrease dramatically with increasing effective overburden pressure, indicating that ambient stress level must be considered in bearing capacity analysis. The results show no evidence of a critical depth below which point resistance is constant; rather point resistance continues to increase with depth, but at a reducing rate.

Finite element methods are being used more and more during design, in cases involving compacted soil‐structure interaction. In general, it is not practical to conduct extensive tests to obtain the compacted soil properties required by the finite element methods during the design phase. Alternatively, if finite element approaches are used for developing design tables, soil properties representative of typical soil types and compaction specifications are required. To provide the needed design soil parameters for a wide variety of soil conditions, laboratory testing is carried out on three soils: a sand, a silt, and a clay. These soils are prepared at density states ranging from loose to 95% of the maximum from the standard compaction tests [American Society for Testing and Materials (ASTM) D698, American Association of State Highway and Transportation Officials (AASHTO) T‐99]. The tests used to obtain the soil parameters are triaxial compression, isotropic compression, and one‐dimensional compression. A consistent set of soil design parameters are obtained by fitting test results to a hyperbolic soil model representing Young's modulus and bulk modulus as functions of stress.

Sensitive soils, in general, are prone to mechanical disturbances while sampling, handling, and testing. This necessitates the prediction of true field behavior. The compressibility response of such soils is typical of having three zones, mechanistically explained as nonparticulate, transitional, and particulate. Such zoning has enabled the development of a simple method to predict the field compressibility response of the sample. The field compression curve with ${\sigma }_{\mathrm{ct}}$ act as the most probable yield stress is considered to reflect 0% disturbance. By a comparison of experimentally determined ${\sigma }_c$ and ${\sigma }_{\mathrm{ct}}\text{,}$ it is possible to estimate the degree of sample disturbance. When the value of ${\sigma }_c$ is closer to ${\sigma }_{\mathrm{ct}}\text{,}$ the sampling disturbance approaches zero. As the value of ${\sigma }_c$ reduces, the degree of sampling disturbance increases. The possibility of using this degree of sample disturbance from compressibility data to obtain other true properties from laboratory results of the sampled specimens has been examined.

It has long been recognized that the one‐dimensional or uniaxial strain response of most soils subjected to high‐intensity transient loads (i.e., blast pulses) differs from the response measured under quasi‐static loading rates. Recent research has suggested that for submillisecond rise times, increases up to 10‐fold in the loading constrained modulus occur for some remolded partially saturated granular soils under undrained conditions. Parallel research has shown that, in contrast, loading‐rate effects can be ignored for a similar granular material tested under nearly identical boundary conditions. Stress‐strain curves from 60 uniaxial strain tests are summarized and presented herein depicting the behavior of three soils (two clean sands and a silty clay) to a variety of loading rates. Loadings are typically carried to 10,000 psi (69 MPa) with times to peak ranging from a few tenths of a millisecond to several minutes. These laboratory test results show that a dramatic increase in the loading constrained modulus does not occur for the rise times examined. Rather, a gradual stiffening occurs as the time to peak pressure decreases. The maximum ratio of the dynamic‐to‐static loading constrained modulus is observed to be about a factor of two for the soils tested. Based upon these test results, a strain‐rate and strain‐level dependent modulus stiffening model is developed. This model is implemented into a one‐dimensional plane wave propagation computer code to predict the results of field tests that are performed using two of the three soils tested in this study. A comparison between the laboratory‐based model predicted behavior and response obtained from the field tests is favorable.

Time‐settlement data of more than 30 different sections of the 55‐km long Bangna‐Bangpakong Highway in Thailand are analyzed. The deformation parameters, namely: undrained modulus, $E_u;$ drained modulus, $E^{\prime };$ and coefficient of consolidation, $C_v\text{,}$ are backfigured from the field performance of the highway embankment, and the following correlations are found: $E_u/S_{\mathrm{uv}}=150;$ $E^{\prime }/S_{\mathrm{uv}}=15;$ and $C_v\left(\mathrm{field}\right)/C_v\left(\mathrm{lab}\right)=26;$ where $S_{\mathrm{uv}}$ is the uncorrected vane shear strength. It was also found that $C_v$ values are overestimated by the method of Asaoka when the during‐construction time‐settlement curve is used, and are best estimated when backfigured from postconstruction performance. It is also found that for construction settlements, the method proposed by Cox, which is a combination of the method of D'Appolonia et al. for immediate settlements and that of Leroueil et al. for consolidation settlements, underpredicts settlements at some sections but yields conservative estimates when considering secondary settlements since the beginning of construction. Prediction of long‐term settlements yields a good estimate for firmer sections using the method of Skempton and Bjerrum, while the method of Asaoka generally underpredicts. The elastic method of Davis and Poulos gives the best estimates of both construction and postconstruction settlements when backfigured parameters are used.

Collapsible soils are susceptible to a large reduction in volume upon wetting. For the soil tested, the sampling process did have an effect on the laboratory test results. The process of driving a Shelby Tube into the collapsible soil deposit caused fracturing which resulted in different consolidation test results than those tests on block samples. For the soil tested, the percent collapse differed between the block samples and the driven Shelby Tube.