No description available.

Necrotic enteritis (NE) caused by Clostridium perfringens type A is an important bacterial enteric disease of global broiler production. However, the dynamic interactions of NE and its predisposing factors are not fully presented by current studies. By using the System Dynamics (SD) Model, the epidemiological changes in susceptible-infected-removed models of NE and avian coccidiosis and their interactions in one or multiple grow-out cycles was established; meanwhile, the growth performance was measured by the average weights of infected and non-infected populations at harvest were estimated. The SD model provided direct and persuasive outcomes of the epidemiology and ecology of NE compared with models using statistical methodology. With interventions on certain predisposing factors of management practices and medication, effects which decreased disease incidence and growth performance were observed; moreover, the leverage points obtained from interventions on certain management practices provided quantitative results which were applicable and useful for improving the broiler production.

<p>Chemical weathering of highly reactive mafic and ultramafic igneous rocks may be a key sink in the global carbon cycle. Understanding how uplift of these rocks during arc-arc and arc-continent collisions through earth history has affected the evolution of global climate, including the onset of icehouse periods, requires improved constraints on the relative sensitivity of their weathering rates to physical erosion vs. climate. If weathering rates depend chiefly on erosion, then tectonic uplift of mafic and ultramafic rocks may have a strongly destabilizing effect on global climate. Conversely, if weathering rates are limited primarily by temperature or runoff, then a negative feedback mechanism between weathering and climate may attenuate the effects of rock uplift. This work characterizes the relationship between chemical weathering rates, physical erosion rates, and climate in tropical, montane watersheds in Puerto Rico that are underlain by volcanic arc rocks and associated ophiolitic serpentinite. Key to this analysis are new constraints on long-term erosion rates on these rocks from cosmogenic Cl-36 produced <em>in situ</em> in magnetite. These cosmogenic erosion rates are paired with classical measurements of stream solute fluxes and sediment geochemistry across runoff gradients to quantify the limits to volcanic arc rock and serpentinite weathering rates. </p> <p><br></p> <p>This work is divided into three chapters. Chapter 2 constrains the altitude scaling behavior of Cl-36 production in magnetite. This allows erosion rates to be determined more accurately in watersheds near sea level in Puerto Rico. Chapter 3 demonstrates that volcanic arc rock weathering rates in the humid tropics are more strongly limited by physical erosion than by climatic factors. However, a positive correlation between erosion and runoff observed in this landscape may enhance the coupling between climate and weathering rates. Chapter 4 finds that, in contrast to volcanic arc rocks, serpentinite weathering is strongly limited by runoff and weakly limited by erosion. These results are presented as empirical power-law relationships that can be readily applied in global carbon cycle modeling.  </p>

In recent years, deep learning has revolutionized computer vision and natural language processing tasks, but the black-box nature of these models poses significant challenges for their interpretability and reliability, especially in critical applications such as healthcare. To address this, attention-based methods have been proposed to enhance the focus and interpretability of deep learning models. In this dissertation, we investigate the effectiveness of attention mechanisms in improving prediction and modeling tasks across different domains. We propose three essays that utilize task-specific designed trainable attention modules in manufacturing, healthcare, and system identification applications. In essay 1, we introduce a novel computer vision tool that tracks the melt pool in X-ray images of laser powder bed fusion using attention modules. In essay 2, we present a mask-guided attention (MGA) classifier for COVID-19 classification on lung CT scan images. The MGA classifier incorporates lesion masks to improve both the accuracy and interpretability of the model, outperforming state-of-the-art models with limited training data. Finally, in essay 3, we propose a Transformer-based model, utilizing self-attention mechanisms, for parameter estimation in system dynamics models that outpaces the conventional system calibration methods. Overall, our results demonstrate the effectiveness of attention-based methods in improving deep learning model performance and reliability in diverse applications. / Doctor of Philosophy / Deep learning, a type of artificial intelligence, has brought significant advancements to tasks like recognizing images or understanding texts. However, the inner workings of these models are often not transparent, which can make it difficult to comprehend and have confidence in their decision-making processes. Transparency is particularly important in areas like healthcare, where understanding why a decision was made can be as crucial as the decision itself. To help with this, we've been exploring an interpretable tool that helps the computer focus on the most important parts of the data, which we call the ``attention module''. Inspired by the human perception system, these modules focus more on certain important details, similar to how our eyes might be drawn to a familiar face in a crowded room. We propose three essays that utilize task-specific attention modules in manufacturing, healthcare, and system identification applications. In essay one, we introduce a computer vision tool that tracks a moving object in a manufacturing X-ray image sequence using attention modules. In the second essay, we discuss a new deep learning model that uses focused attention on lung lesions for more accurate COVID-19 detection on CT scan images, outperforming other top models even with less training data. In essay three, we propose an attention-based deep learning model for faster parameter estimation in system dynamics models. Overall, our research shows that attention-based methods can enhance the performance, transparency, and usability of deep learning models across diverse applications.

<p>The solar system that we live in today bears only a passing resemblance to the solar system that existed 4.5 billion years ago. As our young star shed the gas nebula from which it was born, a disk of dust and rocky bodies emerged in the space between the Sun and Jupiter. Over the next hundred million years, this planetary disk evolved and gave rise to the terrestrial planets of the inner solar system. Clues left behind during this early stage of evolution can be seen in the orbital architecture of the modern planets, the cratering records of rocky bodies, and the signatures of the solar system's secular modes. </p> <p><br></p> <p>Past works in the fields of terrestrial planet accretion and solar system evolution typically do not include collisional fragmentation. While the mechanics of collisional fragmentation are well studied, the incorporation of this processes into simulations of terrestrial planet formation is computationally expensive via traditional methods. For this reason, many works elect to exclude collisional fragmentation entirely, improving computational performance but neglecting a known process that could have played a significant role in the formation of the solar system. In this dissertation, I develop a collisional fragmentation algorithm, called Fraggle, and incorporate it into the n-body symplectic integrator Swiftest SyMBA. Along with performance enhancements and modern programming practices, Swiftest SyMBA with Fraggle is a powerful tool for simulating the formation and evolution of the inner solar system. </p> <p><br></p> <p>In this dissertation, I use Swiftest SyMBA} with Fraggle to study the effect of collisional fragmentation on the accretion and orbital architecture of the terrestrial planets, as well as the cratering record of early Mars. I show that collisional fragmentation is a significant process in the early solar system that creates a spatially heterogeneous and time-dependent population of collisional debris that fluctuates as the solar system evolves. This ever-changing population results in cratering records that are unique across the inner solar system. The work presented in this dissertation highlights the need for independent cratering chronologies to be established for all rocky bodies in the solar system, as well as the need for future models of solar system accretion to include the effects of collisional fragmentation. </p> <p><br></p> <p>While the cratering records and orbits of the terrestrial planets are two means by which to study the solar system's ancient past, analysis of the evolution of the secular modes of the solar system offers a third method. A secular mode arises due to the precession of the orbit of a planet over time. Each body's orbit precesses at a specific fundamental frequency, or mode, that has the power to shape the orbital architecture of the solar system. I show that jumps in the eccentricity of Mars can trigger short-lived power sharing relationships between secular modes, resulting in periods in which the strength and fundamental frequencies of modes fluctuates. While evidence of these past jumps in Mars' eccentricity would likely not be visible today in the secular modes of the inner solar system, the work presented in this dissertation poses additional questions. In particular, questions related to other possible triggers of power sharing relationships, as well as the effects of power sharing relationships on the stability of small bodies during these periods of fluctuation, are particularly compelling.</p> <p><br></p> <p>The work presented in this dissertation contributes to the fields of numerical modeling, solar system evolution, collisional fragmentation, martian cratering, and secular modes and resonances. As a whole, it explores avenues by which we can understand the very earliest period of our solar system's history and develops a model that will allow for continued research in this field. </p>

No description available.

For all business organizations, there comes a time when a change must take place within their eco-system. It consumes a great deal of thought and planning to ensure that the right decision is made as it could alter the entire course of their business for a number of years to come. This change may appear in the form of a brilliant CEO reaching the age of retirement, or an unsuccessful Managing Director being asked to leave before fulfilling the term of her contract. Regardless of the cause, a transition must occur in which a suitable successor is chosen and put into place while minimizing costs, satisfying stakeholders, ensuring that the successor has been adequately prepared for their new position, and minimizing work place gossip, among other things. It is also important to understand how the nature of the business, as well as its financial standing, effects such a transition. Engineering and management principles come together in this study to ensure that organizations going through such a change are on the right course. As the problem of transitional management is not one of concrete values and contains many ambiguous concepts, one way to tackle the problem is by utilizing various industrial engineering methodologies that allow these companies to systematically begin preparing for such a change. By default, organizational strategy has to change, technology is continually being renewed and it becomes very hard for the same leader to constantly implement new and innovative developments. Organizations today have a very poor understanding of where they currently stand and as a result the cause for a company's lack of profitability is often overlooked with time and money being wasted in an attempt to fix something that is not broken. To be able to look at the bigger picture of an organization and from there begin to close in on the main problems causing a negative impact, the Matrix of Change is used and takes in many factors to layout an accurate representation of the direction in which an organization should be headed and how it can continue to grow and remain successful. The Theory of Constraints on the other hand is used here as a step-by-step guide allowing companies to be better organized during times of change. And System Dynamics modeling is where these companies can begin to simulate and solve the dilemma of transitional management using causal loop diagrams and stock and flow diagrams. Through such tools a framework can begin to be developed, one that is valued by corporations and continually reviewed. Several case studies, simulation modeling, and a panel of experts were used in order to demonstrate and validate this framework.

Currently the network centric operation and network centric warfare have generated a new area of research focused on determining how hierarchical organizations composed by human beings and machines make decisions over collaborative environments. One of the most stressful scenarios for these kinds of organizations is the so-called extreme events. This dissertation provides a hybrid simulation methodology based on classical simulation paradigms combined with social network analysis for evaluating and improving the organizational structures and procedures, mainly the incident command systems and plans for facing those extreme events. According to this, we provide a methodology for generating hypotheses and afterwards testing organizational procedures either in real training systems or simulation models with validated data. As long as the organization changes their dyadic relationships dynamically over time, we propose to capture the longitudinal digraph in time and analyze it by means of its adjacency matrix. Thus, by using an object oriented approach, three domains are proposed for better understanding the performance and the surrounding environment of an emergency management organization. System dynamics is used for modeling the critical infrastructure linked to the warning alerts of a given organization at federal, state and local levels. Discrete simulations based on the defined concept of "community of state" enables us to control the complete model. Discrete event simulation allows us to create entities that represent the data and resource flows within the organization. We propose that cognitive models might well be suited in our methodology. For instance, we show how the team performance decays in time, according to the Yerkes-Dodson curve, affecting the measures of performance of the whole organizational system. Accordingly we suggest that the hybrid model could be applied to other types of organizations, such as military peacekeeping operations and joint task forces. Along with providing insight about organizations, the methodology supports the analysis of the "after action review" (AAR), based on collection of data obtained from the command and control systems or the so-called training scenarios. Furthermore, a rich set of mathematical measures arises from the hybrid models such as triad census, dyad census, eigenvalues, utilization, feedback loops, etc., which provides a strong foundation for studying an emergency management organization. Future research will be necessary for analyzing real data and validating the proposed methodology.

Decisions about Modeling and Simulation (M&S) of Complex Systems (CS) need to be evaluated prior to implementation. Discrete Event (DE), System Dynamics (SD), and Agent Based (AB) are three different M&S approaches widely applied to enhance decision-making of complex systems. However, single type M&S approaches can face serious challenges in representing the overall multidimensional nature of CS and may result in the design of oversimplified models excluding important factors. Conceptual frameworks are necessary to offer useful guidance for combining and/or integrating different M&S approaches. Although several hybrid M&S frameworks have been described and are currently deployed, there is limited guidance on when, why and how to combine, and/or integrate DE, SD, and AB approaches. The existing hybrid frameworks focus more on how to deal with specific problems rather than to provide a generic way of applicability to various problem situations. The main aim of this research is to develop a generic framework for Multi-Method Modeling and Simulation of CS, which provides a practical guideline to integrated deployment or combination of DE, SD, and AB M&S methods. The key contributions of this dissertation include: (1) a meta-analysis literature review that identifies criteria and generic types of interaction relationships that are served as a basis for the development of a multi-method modeling and simulation framework; (2) a methodology and a framework that guide the user through the development of multi-method simulation models to solve CS problems; (3) an algorithm that recommends appropriate M&S method(s) based on the user selected criteria for user defined objective(s); (4) the implementation and evaluation of multi method simulation models based on the framework's recommendation in diverse domains; and (5) the comparison of multi-method simulation models created by following the multi-method modeling and simulation framework. It is anticipated that this research will inspire and motivate students, researchers, practitioners and decision makers engaged in M&S to become aware of the benefits of the cross-fertilization of the three key M&S methods.

The most known and widely used methods use cash flows and tangible assets to measure the impact of investments in the organization’s outputs. But in the last decade many newer organizations whose outputs are heavily dependent on information technology utilize knowledge as their main asset. These organizations’ market values lie on the knowledge of its employees and their technological capabilities. In the current technology-based business landscape the value added by assets utilized for generation of outputs cannot be appropriately measured and managed without considering the role that intangible assets and knowledge play in executing processes. The analysis of processes for comparison and decision making based on intangible value added can be accomplished using the knowledge required to execute processes. The measurement of value added by knowledge can provide a more realistic framework for analysis of processes where traditional cost methods are not appropriate, enabling managers to better allocate and control knowledge-based processes. Further consideration of interactions and complexity between proposed process alternatives can yield answers about where and when investments can improve value-added while dynamically providing higher returns on investment