A system of interest usually consists of some unknown model parameters that affect its output. System diagnosis estimates these model parameters and track their evolution if the system is time-dependent. Subsequently, the prognosis predicts the system output at future inputs. An important challenge in diagnosis and prognosis is the presence of various uncertainty sources, such as natural variability, inadequate data, and approximate models. The first challenge is how to integrate the contributions of the different uncertainty sources towards the overall prediction uncertainty; dimension reduction is another challenge in the case of a large number of uncertainty sources; other challenges includes test design, computational efficiency, etc. This dissertation uses the Bayesian network and variance sensitivity analysis as major mathematical tools and develops multiple innovations to solve the aforementioned challenges in system diagnosis and prognosis. Regarding sensitivity analysis, this dissertation proposes a framework to incorporate both aleatory and epistemic uncertainty, and a new sample-based algorithm to significantly improve the computational efficiency. This leads to sensitivity analysis of the Bayesian network for dimension reduction. Regarding uncertainty integration, this dissertation proposes a roll-up method to incorporate the results from multiple uncertainty quantification activities, and a sensitivity-based optimization approach for test design. A dynamic Bayesian network is utilized for the diagnosis and prognosis of time-dependent systems, and illustrated with an aircraft wing digital model to monitor the health status of the wing. A fast Bayesian inference algorithm is also proposed to improve the computational efficiency, thus enabling real-time diagnosis and prognosis for decision support. In sum, this dissertation covers multiple topics in uncertainty quantification and system health monitoring, and the proposed methodologies/algorithms provide valuable breakthroughs for comprehensive uncertainty integration and higher computational efficiency without compromising accuracy.

<p> Traffic congestion is a daily occurrence in urban highway networks worldwide. It is not possible, however, for society to build its way out of congestion; rather, smarter roads and vehicles are needed. While the development of a smarter transportation system is underway, full implementation is years or decades from now. Yet, some of the sensing technology needed for smarter vehicles is already widely deployed in the form of smart phones. This thesis develops a novel method for recognizing traffic congestion using an artificially intelligent heuristic that could be implemented in a smart phone application or embedded system. Its goal is to provide intelligent feedback to a driver or autonomous vehicle control system to counteract stop-and-go traffic, a defining feature of urban highway congestion. Evaluation of the method indicates that a specific condition during stop-and-go traffic can be recognized accurately. A driver or control system acting upon feedback provided by the artificially intelligent system can improve traffic flow on the roadway by 1% to 3.5% over the course of the test duration.</p>

Composite materials are attractive for aerospace vehicles due to their low weight and high performance; however, predictive computational models are needed to lower the costs associated with their design and certification process. Furthermore, because aircraft structures are built up from many components fastened together, it is critical to understand and predict the mechanical behavior bolted joints introduce. The goal of this thesis is to advance the state of the art in fiber reinforced polymer composite bolted joint computational modeling and prediction using the multiscale reduced order computational model known as the Eigendeformation-based Homogenization Method (EHM). Preliminary blind predictions were performed for laminated [44/44/11] IM7/977-3 composite specimens in single-lap straight hole and countersunk bolt joints under static tension tests. To address stiffness and post-peak behavioral issues, an in-depth blind prediction study was conducted for open hole, filled hole, and single shear bearing configurations of countersunk laminated [44/44/11] IM7/977-3 specimens under static tension and compression tests. EHM performed very well in the open and filled hole cases, and the single shear bearing results show promise of EHM as a reliable choice for composite bolted joint damage analysis.

<p> Industry practitioners and researchers recognize project scope definition as a factor of project success in terms of cost and schedule. The Construction Industry Institute developed the Project Definition Rating Index (PDRI) as a tool to aid in the effectiveness of scope development. The Air Force has adopted use of this tool, though has yet to validate its effectiveness empirically. The objective of this study is to provide that empirical validation by comparing the cost, schedule, and budget estimate performance metrics of Air Force military construction (MILCON) projects that used the PDRI against those that did not. Project data for 263 (100 PDRI and 163 non-PDRI) MILCON projects worth $3.9 Billion were analyzed. The projects that used the PDRI performed better on all three metrics, with statistically significant results on both cost and schedule growth. This study provides empirical evidence of how the use of formal scope definition tools can improve performance for Air Force MILCON projects. When compared to previous research, the study also contributes to a broader understanding of scope definition in the design and construction industry.</p>

No description available.

No description available.

No description available.

<p> As population and traffic increase, improvements in the paving industry including roadways, parking lots, and sidewalks are desired. Pavements have a key effect on the quantity and quality of rainwater runoff and reserved ground water. More so, regulations towards constructing and paving these new areas are becoming stricter especially in urban areas where available land is becoming limited. Permeable pavements are more common than ever, and are a solution to rainwater runoff, recharging ground water, and reducing the costs associated with treating storm water. </p><p> Permeable pavements are available alternative to conventional pavements and are becoming more applicable and more widespread. Unlike conventional pavements, permeable pavements have major obstacles to acceptance and use since there is limited research on these types of pavements, with only a handful of guidelines and properties for users to follow. </p><p> This research incorporates two technical concepts on two topic areas. The first topic can be used to assist decision makers, planners, and owners in selecting their pavement type for their intended use by analyzing and comparing the life-cycle cost for four pavement types. Two of these are conventional impermeable pavements: Hot-Mix Asphalt and Portland Cement Concrete and two are permeable pavements: Porous Asphalt and Pervious Concrete. This first topic area will also summarize tables of advantages and limitations for each pavement type. The second topic area of this paper will help initiate guidelines on how to design, mix, batch, place, cure, and test pervious concrete.</p>

<p> Recent earthquakes devastated lives and destroyed a great stock of buildings. As a result, the earthquake-impacted regions incurred huge business and operation interruption losses. To minimize the business interruption losses through Performance-Based Seismic Design, there is an obvious need for a validated downtime model that would cover a large spectrum of building sizes and types. Building downtime consists of securing finances, mobilizing contractors, engineers and supplies, and the time to perform the actual repair, i.e., repair time. This study focuses on developing a model to characterize the repair time contribution to the downtime as an extension to FEMA P-58 Loss Assessment Methodology. The proposed repair time model utilizes the Critical Path Method for repair scheduling and realistic labor allocations that are based on the amount and severity of building damage. The model is validated on a significant sample of data collected through case studies from previous earthquakes, interviews with contractors, engineers, and inspectors. The proposed model also has a capability of scheduling resources to meet resource limitations that can either come from labor congestions or from a surge in demands following a disaster. The proposed resource scheduling method provides an efficient way of reducing the number of workers during labor congestions while minimizing its effect on the project duration. The final outcome is a realistic estimation of repair time associated with an earthquake.</p>

This dissertation presents the formulation and implementation of the variational multiscale enrichment computational framework for scale inseparable multiscale modeling of structures subjected to extreme environments. In the presence of structures with elasto-viscoplastically behaved heterogeneous materials, the framework includes the variational multiscale enrichment (VME) method, the reduced order variational multiscale enrichment (ROVME) method for mechanical and thermo-mechanical problems, and the hybrid integration for reduced order variational multiscale enrichment (HROVME) method. First, the variational multiscale enrichment method for elasto-viscoplastic problems is developed for the scale inseparable multiscale modeling. VME is a global-local approach that allows accurate fine scale representation at small subdomains whereas the response within far-fields is idealized using a coarse scale representation. The scale inseparable character is represented by the relatively insignificant scale size difference and strong coupling effect between the scales. A one-parameter family of mixed boundary conditions that range from Dirichlet to Neumann is employed to study the effect of the choice of boundary conditions at the fine scale on accuracy. Second, the reduced order variational multiscale enrichment method for elasto-viscoplastic problems is developed to improve the computational efficiency of the VME method. By eliminating the requirement of direct fine scale discretization and repetitive evaluation of the microscale equilibrium state, the computational effort associated with the VME method is significantly reduced. Third, the reduced order variational multiscale enrichment method for coupled thermo-mechanical problems is presented which extends the ROVME method to model structures with temperature sensitive constituent properties. The temperature-dependent coefficient tensors of the reduced order approach are approximated in an efficient manner, retaining the computational efficiency of the reduced order model in the presence of spatial/temporal temperature variations. Last, the hybrid integration for reduced order variational multiscale enrichment method is developed to further improve the computational efficiency of the proposed framework. Considering the coupled transport-thermo-mechanical effects, it employs the key ideas of the ROVME and the computational homogenization approaches to directionally consider scale separation within the structures. The HROVME method also extends the ROVME approach to microstructures with periodic boundary conditions and improves the stability of the ROVME method by avoiding the potential hourglass modes. Numerical verifications are performed to demonstrate the high accuracy, computational efficiency and capability of the proposed computational framework.