Conventional methods for visual assessment of civil infrastructures have certain limitations, such as subjectivity of the collected data, long inspection time, and high cost of labor. Although some new technologies (i.e. robotic techniques) that are currently in practice can collect objective, quantified data, the inspector's own expertise is still critical in many instances since these technologies are not designed to work interactively with human inspector. This study aims to create a smart, human-centered method that offers significant contributions to infrastructure inspection, maintenance, management practice, and safety for the bridge owners. By developing a smart Mixed Reality (MR) framework, which can be integrated into a wearable holographic headset device, a bridge inspector, for example, can automatically analyze a certain defect such as a crack that he or she sees on an element, display its dimension information in real-time along with the condition state. Such systems can potentially decrease the time and cost of infrastructure inspections by accelerating essential tasks of the inspector such as defect measurement, condition assessment and data processing to management systems. The human centered artificial intelligence (AI) will help the inspector collect more quantified and objective data while incorporating inspector's professional judgment. This study explains in detail the described system and related methodologies of implementing attention guided semi-supervised deep learning into mixed reality technology, which interacts with the human inspector during assessment. Thereby, the inspector and the AI will collaborate/communicate for improved visual inspection.

This research aims to utilize available technologies to assess the structural performance of prestressed concrete bridges. The structural performance is investigated at the local level in the form of delamination and spalling and at the global level in the form of strength and serviceability load rating and reliability index. At the local level, Infrared scanning combined with high-definition image scanning is utilized to provide a rapid condition assessment of the bridge components (deck, superstructure, and substructure). Two scans are performed, a year apart. The results of the two scans are compared to each other and to the bridge element level inspection report. A new approach is developed to identify areas where further evaluation is recommended. At the global level, A novel approach is introduced to assess the impact of girders' deterioration on the load rating, component reliability index, and system reliability index of the bridge. The impact of repairs to the girders is also assessed while accounting for uncertainties in the loading, capacity terms, and repair procedures. Computer vision-based load test, validated and supplemented by strain gauge and potentiometers sensors, is utilized to measure deflection under static and dynamic loading conditions. The effect of field-derived distribution factors and dynamic impact factors is investigated for the load rating and component and system reliability indices. The findings of this investigation are used to develop a decision tree for recommending the utilization computer vision-based load test to explore possible increases in load rating or target component reliability.

Pile driving, a commonly used method for installing deep foundations, has gained prominence as a foundation solution to transfer structural loads to deep competent strata. However, this method of installation can generate noise, ground vibrations, and deformations. These effects pose risks to adjacent structures and buried utilities, jeopardizing the safety and serviceability of urban infrastructure. Researchers and public and private agencies have proposed many vibration limit criteria to avoid damage to infrastructure. However, these criteria for construction vibrations are not linked to the ground densification associated with repetitive and cumulative loadings in sandy soils. This dissertation focuses on developing a prediction semi-empirical model to determine ground deformations and vibrations induced by impact pile driving in granular soil deposits. Field data of ground deformations and vibrations were collected by monitoring 13 project sites in Central Florida during the installation of precast prestressed concrete piles using impact hammers. A continuous pile driving modeling approach, in which the pile is driven without any interruption to a final target depth, was coupled with an Updated Lagrangian approach in the numerical framework. An advanced constitutive soil model (i.e., hypoplasticity for sands enhanced with the intergranular strain concept) capable of reproducing changes in the soil void ratio during pile driving was adopted by computationally matching the nonlinear behavior of the granular layer with published shear modulus degradation curves. A critical highly disturbed zone was defined due to the computed soil liquefaction. The developed prediction model is validated with field data, previously published vibration attenuation curves, and vibration-induced ground surface settlement prediction methods in terms of its ability to estimate ground vibrations and deformations induced by impact pile driving in this study. Semi-empirical equations and charts are proposed using a combination of field measurements and numerical analyses to consider the following variables for the ground response due to impact pile driving operations: (1) rated energy of the hammer, (2) scaled distance from the pile, (3) pre-drilling depth, and (4) soil relative void ratio, which is related to relative density. The findings indicated that large ground deformations can occur even in cases where vibration levels (i.e., peak particle velocities) do not exceed the vibration limits.

This thesis investigated the potential advantages and disadvantages of high performance structures by comparing the financial and environmental impacts of a performance based four-story office building to one designed to meet minimum code-level requirements. To generate a comparison, the lateral system of a four-story structure utilizing buckling restrained braced frames was designed to meet code-level requirements per the American Society of Civil Engineers’ Minimum Design Loads for Buildings and Other Structures (ASCE 7-05) and again to meet the immediate occupancy criteria defined by ASCE 41-06 Seismic Rehabilitation of Existing Buildings. The following was then performed: Test the structural performance of both buildings using simulated code-level and maximum considered earthquakes Develop construction costs of both structures using RSMeans Square Foot Cost and Construction Cost Data Determine the financial benefit associated with the upgraded structure by subjecting both structures to a suite of earthquakes Calculate the carbon footprint generated during each building’s construction. The final project costs for the code level and immediate occupancy structures were $27.43 million and $27.93 million respectively, resulting in an upgrade cost of $500,000 or roughly 1.8% of the overall project cost. The upgrade cost was then input in FEMA’s Benefit-Cost Analysis, where it found the upgrade cost resulted in an annual savings ranging from $43,000 to $98,000 over the building’s 50-year life cycle. The carbon footprints were generated using BuildingScope, which relies on volumetric quantities of construction materials. The final models resulted in a carbon footprint of 7890 CO2 eq and 7940 CO2 eq for the code level and immediate occupancy structures respectively, showing favor for the structure utilizing fewer materials. Although the additional materials used in the immediate occupancy structure resulted in a slightly larger carbon footprint, the added capacity will decrease damages, resulting in an overall reduction of energy generated during the building’s life cycle.

Post-earthquake reconnaissance and results of previously conducted experiments show that stiffness and strength deterioration of beam-column joints can have a detrimental effect on the integrity and vulnerability of reinforced concrete frame structures, especially in older buildings in developing countries. As a result, there is a need to develop efficient structural evaluation techniques that are capable of accurately estimating the strength and deformability of existing buildings to facilitate the development of safer, simpler, and lower cost retrofit solutions and thus contributing to risk mitigation. The current research is part of a general effort that is being carried out at the University of Sheffield to quantify and develop strategies for the mitigation of seismic risk in developing countries. The primary aim of this work is to improve the current understanding of the seismic behaviour of deficient exterior reinforce concrete beam-column joints. Seven full-scale isolated exterior beam-column joints were tested under quasi-static cyclic loading to investigate and quantify the effects of using different types of beam reinforcement anchorages and low column axial loads on the seismic shear performance of exterior beam-column joints with no shear reinforcement. Contrary to what is reported in the literature, the test results show that increasing the column axial load even at very low levels «O.2f'oAg,) can enhance the joint shear strength of deficient exterior joints (exhibiting pure shear failure) by up to 15%. The test results also show that, for the same joint panel geometry and column axial load, the type of beam anchorage detail, whether it is a straight bar, long or short hook, can influence the joint shear strength by up to 34%. A new analytical model that predicts the shear strength of deficient exterior beam-column joints in both loading directions and takes into account the column axial load and bond conditions within the joint is developed. The model predicts with good accuracy the strength of the tested specimens in addition to other specimens reported by other researchers. Furthermore, a springbased exterior beam-column joint model for finite element analysis of deficient RC frames is proposed. The model development includes a joint shear stress-strain constitutive model based on the developed strength model. The simulated response using the proposed model shows good agreement with the experimentally observed response.

Roller Compacted Concrete (RCC) is a mixture of well graded aggregates, cement and water. It is placed with a high compaction asphalt type paver and compacted to high density by vibratory rollers to provide a high strength and durable pavement structure. RCC requires no formwork, surface finishing, dowelled joints or reinforcement. These characteristics make RCC simple, fast and economical. However, it also presents difficulties with high-speed applications related to surface texture and surface evenness. These difficulties have so far restricted the use of RCC to the lower layers of normal roads. A two-layer RCC pavement system is a type of composite pavement consisting of two concrete layers. The two layers are paved in either a “wet-on-wet” technique or ”wet-on-dry” technique. The bottom layer serves as the main bending-resistant component of the composite slab, while the top lift is generally constructed with higher-quality constituent materials for improved surface characteristics such as noise and skid resistance. The aim of this research is to evaluate and design two-layer RCC systems with different aggregate sizes and types and different placement conditions in order to expand the application of RCC in pavements. Mechanical properties, bond strength properties, durability characteristics, surface properties, fatigue damage, joint deterioration and pavement design are the main. A range of testing equipment, methodologies and tools have been used in this investigation. The findings of this study showed that a two-layer RCC system can achieved good strength and stiffness for each of the mixtures in the two layers. Also, the inter-layer bond was found to be strong when the two layers were placed within one hour, but weaker when the upper layer was placed three hours after the lower layer. Moreover, the durability of the two-layer RCC system was found to be acceptable, especially when the upper layer was placed within an hour of the lower layer. The surface characteristics for the upper layer of RCC showed that the minimum requirement for skid resistance and texture depth have been achieved. However, it is suggested that further investigation is needed, particularly into the evenness of RCC. The investigation into the effect of dynamic load on the two-layer RCC system demonstrated a good fatigue strength for each RCC mixture and for the two layers together, compared to conventional concrete pavements. Also, the results of load transfer stiffness and joint deterioration showed acceptable performance with regard to crack or joint width, shear stress and placement conditions. The effect of other parameters such as moisture and differential temperature requires a separate investigation and is recommended for future work. The results of the design and analysis of two-layer RCC using KENSLAB, a finite element program, indicted that RCC could perform successfully in pavements with a long service life. In conclusion, the results obtained show that the two-layer RCC are a valid alternative for pavements. On one hand, the use of a harder and more resistance aggregate at the top layer guarantees higher skid resistance and durability, while limiting for the use of high quality aggregate. On the other hand, the results show that adequate construction techniques can alleviate the problems arising from the lack bond between layers.

An extensive amount of works have been carried out to develop the current understanding in punching shear mechanism noted in reinforced concrete slabs. However, despite the increasing popularity of waffle slabs, the current understanding about punching behaviour is mainly focused on solid flat slabs, and only limited amount of works have been carried out on waffle slabs and in the presence of biaxial moment. Thus, there is a need to carry out a research in this area to aid the understanding about punching mechanism of waffle slabs in the presence of biaxial moment for the internal column and edge column connections. The experimental work carried out in this research included destructive testing of thirty-eight 1/10th scale model waffle slab specimens, which consists of fifteen internal column slabs and twenty-three edge column slabs. The main variables were, for the internal column slab, the principle angles of biaxial moment transfer, the column eccentricity, the column orientation and the size of solid sections, and for the edge column slab, the principle angles of biaxial moment transfer, the column eccentricity, the column location and the size of solid sections. From the experimental investigations, three distinct failure mechanisms were observed: the concentric punching at internal column mechanism; the eccentric punching at internal column mechanism; and the edge punching mechanism. In general, the observed punching shear failure mechanisms of waffle slabs were found identical to solid flat slabs; but the punching shear capacities reduced due to some losses in potential failure surface within the waffle section. The principle angle of biaxial moment transfer was found varying the shear surface area that was being mobilized, thus affecting the punching capacity of the slabs. An analytical study was carried out, using an upper-bound plastic model, to simulate the observed punching shear mechanisms, and hence, to predict the punching capacity of the slabs. A theoretical model was developed for each of the identified failure mechanism. In addition, three design models based on the current UK code, Eurocode 2, have been developed. In all cases, these models have achieved good agreements with the test results.

The ability to predict the effect of dimension and thickness variability on the dynamic response of realistically uncertain engineering structures is examined in this thesis. Initially, available methods for predicting key response statistics and probabilities, for both low and high frequencies are examined to establish their strengths and limitations for specified levels of random dimension variability. For low frequency applications, the ability of Direct Integration (DI) and the First-Order Reliability Method (FORM) to predict exceedance probability is examined. For high frequency applications, the ability of the methods of Statistical Energy Analysis (SEA) and DI to predict the mean and standard deviation of the energy response is examined. The use of Extreme Value (EV) theory is included as a way to bound responses using simulated or measured responses. The strengths and limitations of Monte Carlo simulation methods are explored for both low and high frequency responses of randomly uncertain structures using both simple mode superposition plate-structure solutions and (commercially available) finite element solutions for coupled plate structures. To address, without the need to undertake expensive Monte Carlo simulation, the problem of predicting response bounds for structures with varying levels of uncertainty, a novel DI-EV method is developed and examined. It is tested first on a simple plate structure, then on a coupled plate structure, with low-level and high-level random dimension and thickness uncertainty. In addition, the method is compared with the SEA-EV method. The thesis shows that the results from the existing SEA-EV bounding approach gives good bounds only at particular frequencies and mainly for low levels of dimension variability. In contrast, the proposed DI-EV bounding approach compare extremely well with Monte Carlo simulations, which is not only at all frequencies but also with both low-level and high-level uncertainties, for simple and coupled plate structures with dimension and thickness variation.

This thesis is concerned with efficient uncertainty identification (UI) – namely the nonlinear inverse problem of establishing specific statistical properties of an uncertain structure from a practically-limited supply of low-frequency dynamic response information. An established UI approach (published in 2005) which uses Maximum Likelihood Estimation (MLE) and the Perturbation Method of uncertainty propagation is adopted for the study using (for the first time) mode shape information rather than just natural or resonant frequencies. The thesis develops a method based on the use of selected coefficients in a generalized displacement model i.e. a weighted series of spatially-continuous multiply-differentiable base functions to approximate the structural free-vibration response of an uncertain structure. The focus is placed on the estimation (from relatively small data sets) of the statistical properties of the location of an attached point-mass with normally-distributed position. Simulated data for uncertain point-mass-loaded linear beam and plate structures is initially used to test the method making use of as much exact or closed-form differentiable information as possible to obtain frequencies and mode shapes. In the case of plate structures, extensive use is made of the Rayleigh Ritz method to generate the required response coefficients. This is shown to have significant advantages over alternatives such as the Finite Element method. The approach developed for use with free vibration information is then tested on measured experimental data obtained from an acoustically-forced clamped plate. Structural displacement measurements are taken from the plate using Vibromap 1000, a commercially-available ESPI-based holomodal measurement system capable of wide-field vibration response observation in real-time, or quantitative displacement response measurement. The thesis shows that the developed uncertainty identification method works well for beams and plates using simulated free-vibration data

Thesis--M. Sc.(Eng.), University of Hong Kong. / Mimeographed.