Operations Research, Systems Engineering and Industrial Engineering Commons^™

On Intrinsic Dimensionality Of Data Sets And Neural Networks, Ori Chachmo

Theses and Dissertations

The concept of Intrinsic Dimensionality (ID) is of special interest in the field of Neural Networks (NNs) since it promotes both (a) a deeper understanding of the underlying mechanisms, and (b) embraces parsimonious modeling (that is, building the right-sized model for the task) with associated benefits to processing speed and storage requirements. This thesis explores the concept of ID via two separate, but related, questions. First, we study the potential of NN ID prediction by exploiting easily obtained quantities measured on the data. We then explore NN ID as an independent concept by comparing the results of different methods for …

2033 Digital Modernization At Usmepcom: A Strategic Analysis Of Future Military Applicant Processing, William A. Clay

Theses and Dissertations

This research examines the projected 2033 applicant processing scenario considering the digital modernization efforts of the United States Military Entrance Processing Command (USMEPCOM). The study evaluates the necessary modifications to current processes, with a particular focus on the influence of two key information technology systems, the MEPCOM Integrated Resource System (MIRS) 1.1 and the Military Health System (MHS) Genesis, on manpower at a Military Entrance Processing Station (MEPS). In doing so, the study establishes baseline processing metrics for assessing these impacts. By utilizing discrete event simulation modeling and leveraging current literature, the study proposes strategies for incorporating technological advancements into …

Evaluation Of Vtol-Capable Cargo Uavs For Dispersible Airfield Logistics In The Usindopacom Aor, Maria C. Chedzoy

Theses and Dissertations

This research models and analyzes the ability of commercial cargo UAVs to rapidly evacuate logistics from an airfield to proximal, outlying destinations, particularly in the USINDOPACOM AOR. This is a tenet of Agile Combat Employment by the USAF, which seeks to mitigate the effect of kinetic threats by near-peer adversaries. The analysis sets forth a binary linear program to minimize the total time to evacuate a fixed amount of logistics from an airfield. Parameters include the cargo UAV with its performance specifications, number of cargo loading points at the airfield, number of destinations for cargo evacuation, and subset of destinations …

Simulating Human-Autonomous Aircraft Teams In An Anti-Access Area Denial (A2ad) Environment, Michael Kaminski

Theses and Dissertations

The role of autonomy has evolved recently, demanding tighter integration between human and autonomous systems, particularly in highly contested A2AD environments. Near-peer adversaries have modernized their integrated air defense systems (IADS), diminishing the current advantages of the United States Air Force. To regain air dominance, efforts like the Collaborative Combat Aircraft (CCA) program are underway, aiming to deploy unmanned autonomous alongside manned next-generation fighter aircraft. This research assesses various operational concepts, focusing on autonomous tactics post-manned fighter loss, strike timing of independent teams, and weapon configuration observability. Using the Advanced Framework for Simulation, Integration and Modeling (AFSIM), an agent-based model …

Techniques For Addressing Extreme Class Imbalance For Artificial Neural Networks Training, Colin W. Foley

Theses and Dissertations

This research examined the class imbalance problem while training convolutional neural networks (CNN) by applying different techniques to combat this common issue. This research used a modified CIFAR-10 dataset along with a curated aerial image dataset. Methods covered included undersampling, oversampling, synthetic minority oversampling technique, Edited Nearest Neighbors and combinations of the aforementioned methods. This research found that undersampling methods tended to outperform oversampling methods. While undersampling methods showed a decrease in overall accuracy, the increase in minority class prediction performance was promising enough to warrant further investigation.

U.S. Army Cadet Command Branch Prediction Model, Daniel M. Krizan

Theses and Dissertations

The current system for providing US Army ROTC cadets their branches leaves significant uncertainty until the final pronouncement of branch assigned. This uncertainty can be alleviated by providing a prediction model for cadets to input personal data and desired branch to identify likelihood of receiving the request. This thesis produces a machine learning model capable of producing branch prediction for cadets.

Automated Image Registration For Titanium Aircraft Components Via Resolution-Robust Parallel Neural Networks, Paige T. Luebbering

Theses and Dissertations

Titanium alloys are vital to the structural integrity of military and commercial aircraft, comprising numerous critical components. These components are composed of microtexture regions (MTRs) that, at a specific size and orientation, can lead to aircraft failure. Existing MTR testing methods, such as Electron Backscatter Diffraction, often fall short in effectively detecting these MTRs without causing damage to the component. Addressing this gap, this thesis develops a Parallel Convolutional Neural Network (CNN) model tailored for multi-resolution image registration of Polarized Light Microscopy (PLM) images to enhance MTR identification in a non-invasive manner. The findings reveal a significant enhancement in the …

Applying Digital Engineering To Defense Acquisitions Through Model-Based Systems Engineering And Discrete Event Simulation, Michael T. Shutlock

Theses and Dissertations

The U.S. Department of Defense (DoD) faces a critical challenge as acquisition professionals strive to grasp the intricacies of the acquisition process. This study proposes an innovative approach to cultivate a more informed and capable acquisition workforce through the integration of digital engineering, specifically Model-Based Systems Engineering (MBSE) and Discrete Event Simulation (DES) toolsets. Embracing digital transformation with MBSE provides a comprehensive understanding, implementing step-by-step procedures with an interface designed to handle vast amounts of information differently. The merging of MBSE's visual modeling with DES's dynamic simulation offers a holistic view, empowering acquisition professionals with robust planning and risk management …

Assessing Adoption Barriers Of Sustainable Packaging In Egypt, Carol Ramses Morgan

Theses and Dissertations

Sustainable packaging has become an essential part of business decisions and corporate directions. With the rise of environmental damages due to improper waste management and unsustainable practices, businesses have a major responsibility to analyze their products’ life cycles and redesign them with sustainability in mind. Applying sustainable packaging could save companies large amounts of resources, therefore cutting costs, while also achieving the legal and social duty as a corporation towards society and the environment. Many developing countries, with specific focus on Egypt, have recently focused on legislative and corporate decisions in order to encourage more sustainable practices. Egypt’s new Waste …

Investigation Of Machine Learning Driven Adaptive Control Strategy For Heat Exchanger Under Varying Fouling Conditions, Reshma Madhu Pk

Theses and Dissertations

Fouling is an undesirable and inevitable phenomenon that affects Heat Exchanger (HE) dynamics, increases maintenance cost (>50%), and loss of production. Conventional fouling prediction and control techniques demand an in-depth understanding of the HE-specific fouling dynamics. Hence, machine learning emerges as an ideal tool for learning the fouling process from the available measurement dataset.

The proposed work aims to design a machine learning-driven adaptive control technique for an industrial HE, subjected to varying fouling conditions. Naphtha cooler, an industrial HE utilized in petroleum refineries, employs cooling water to lower Naphtha temperature, but the high impurities in the recycled cooling …

Integrated Computer-Aided Design, Experimentation, And Optimization Approach For Perovskites And Petroleum Packaging Processes, Swapana Subbarao Jerpoth

Theses and Dissertations

According to the World Economic Forum report, the U.S. currently has an energy efficiency of just 30%, thus illustrating the potential scope and need for efficiency enhancement and waste minimization. In the U.S. energy sector, petroleum and solar energy are the two key pillars that have the potential to create research opportunities for transition to a cleaner, greener, and sustainable future. In this research endeavor, the focus is on two pivotal areas: (i) Computer-aided perovskite solar cell synthesis; and (ii) Optimization of flow processes through multiproduct petroleum pipelines. In the area of perovskite synthesis, the emphasis is on the enhancement …

Developing Machine Learning And Time-Series Analysis Methods With Applications In Diverse Fields, Muhammed Aljifri

Theses and Dissertations

This dissertation introduces methodologies that combine machine learning models with time-series analysis to tackle data analysis challenges in varied fields. The first study enhances the traditional cumulative sum control charts with machine learning models to leverage their predictive power for better detection of process shifts, applying this advanced control chart to monitor hospital readmission rates. The second project develops multi-layer models for predicting chemical concentrations from ultraviolet-visible spectroscopy data, specifically addressing the challenge of analyzing chemicals with a wide range of concentrations. The third study presents a new method for detecting multiple changepoints in autocorrelated ordinal time series, using the …

Efficient Connectivity Management And Path Planning For Iot And Uav Networks, Amirahmad Chapnevis

Theses and Dissertations

This dissertation explores how to better manage resources in mobile networks, especially for enhancing the performance of Unmanned Aerial Vehicles (UAV)-supported IoT networks. We explored ways to set up a flexible communication architecture that can handle large IoT deployments by making good use of mobile core network resources like bearers and data paths. We developed strategies that meet the needs of IoT networks and enhance network performance. We also developed and tested a system that combines traffic from several mobile devices that use the same user identity and network resources within the core mobile network. We used everyday smartphones, SIM …

A Systems Approach To Process Design And Sustainability - Synergy Via Pollution Prevention, Control, And Source Reduction, Emmanuel Apau Aboagye

Theses and Dissertations

Historically, process design prioritized efficiency and profitability, often overlooking environmental and societal implications. However, given the global challenges like climate change and resource scarcity, there is a growing emphasis on embedding sustainability into process design. Adopting a systems-oriented approach provides a comprehensive view, spanning from raw material acquisition to end-of-life product management. Such an approach not only identifies potential sustainability challenges but ensures that solutions foster both environmental responsibility and economic viability. In this study, a comprehensive framework for designing industrial systems is introduced, aiming to encompass the entire lifecycle impacts of chemical processes. The research initially delves into two …

An Investigation Into Applications Of Canonical Polyadic Decomposition & Ensemble Learning In Forecasting Thermal Data Streams In Direct Laser Deposition Processes, Jonathan Storey

Theses and Dissertations

Additive manufacturing (AM) is a process of creating objects from 3D model data by adding layers of material. AM technologies present several advantages compared to traditional manufacturing technologies, such as producing less material waste and being capable of producing parts with greater geometric complexity. However, deficiencies in the printing process due to high process uncertainty can affect the microstructural properties of a fabricated part leading to defects. In metal AM, previous studies have linked defects in parts with melt pool temperature fluctuations, with the size of the melt pool and the scan pattern being key factors associated with part defects. …

Gaze Tracking Embedded Collaborative Robots For Automated Metrology And Reverse Engineering, Sachithra H. Karunathilake

Theses and Dissertations

Conventional geometric metrology, or three-dimensional (3D) scanning, and reverse engineering heavily rely on the experience of the operators. With an increasing need for automation, robot arms have been adopted for this task. However, due to the large variety of parts and designs, automated path planning could provide a scanning solution that may overlook the critical area, which could potentially deteriorate the scan results. This study explores the integration of collaborative robotics (cobots) with eye tracking technology to improve the autonomous 3D scanning process. The primary objective of this study is to enhance the accuracy and efficiency of cobots in …

A System Theoretic Process Analysis Framework And Model Based Approach For Resilient Space Architecture Design, Eric T. Sommer

Theses and Dissertations

As the capabilities provided by space-based systems offer significant contributions toward defense applications, potential adversaries stand to gain significant value in disrupting them. Therefore, the United States must pursue the development and operation of resilient space architectures, capable of delivering capabilities in the face of disruptions. To support this development, systems engineering methods require innovation to effectively ensure design of complex space architectures to meet their objectives. This thesis recommends and demonstrates a System-Theoretic Process Analysis (STPA) framework to qualitatively analyze space architectures. The analysis outputs identify design considerations, requirements, and constraints required for resilience. To enable a model-based systems …

Task Optimization Utilizing Digital Transformation Concepts - Automation Project Execution Via Agile Methodology, Anthony Steven Maiello

Theses and Dissertations

Task Optimization via the use of automated process improvements is becoming more widespread as more industries lean into the concepts surrounding digital transformation. This shift also necessitates a complementary adaptation in project management methodologies to support the rapid and ever-changing environment, requirements, and innovations. This thesis examines the effectiveness of Agile methodology in managing digital automation projects, with a specific focus placed on process improvements with systems engineering. It accomplished this by contrasting the original model, designed and derived utilizing traditional project management techniques, with the proposed model which is a direct result of the application of Agile project practices. …

Design Space Visualization And Exploration For Many Goal Problems Under Uncertainity, Niharika Balaji

Theses and Dissertations

ABSTRACT

Designing a complex engineered system is challenging due to many conflicting goals, uncertainties, and multiple interactions. Traditional optimization approaches often yield single-point solutions, which may not be suitable for early design stages due to their susceptibility to changes in conditions and uncertainties. To address this challenge, a satisficing approach is employed. This approach enables designers to effectively navigate the design space and identify satisficing solutions that balance conflicting goals in the face of uncertainties and changes in conditions. From a systems design perspective, we view design as an iterative process that involves making informed decisions based on available information …

An Improved Saliency Map With Trustworthiness For Localizing Abnormalities In Medical Imaging, Nolan C. Skelly

Theses and Dissertations

Saliency maps are a widely used methodology to make deep learning models more interpretable. They provide post-hoc explanations through identification of the most pertinent areas of an input medical image. These techniques have been assessed based on 1) localization utility, 2) sensitivity to model weight randomization, 3) repeatability, and 4) reproducibility. Ten saliency map techniques will be tested, Grad, Smooth Grad, Integrated Gradients, Smooth Integrated Gradients, XRAI, Grad-CAM, Guided Backpropagation, Guided GradCAM, GradCAM++, and ScoreCAM. The neural networks used to predict and read medical information require a reliable solution to provide medical practitioners intelligible results. Using the information of two, …

Federated Active Learning For Network Intrusion Detection, Matthew D. R. Sauer

Theses and Dissertations

This thesis addresses challenges with detecting attacks on computer networks within a Federated Learning (FL) framework, when labeled instances are few. We explore the integration of active learning (AL) and semi-supervised learning (SSL). AL efficiently uses data that would otherwise be wasted or require substantial time for labeling. SSL provides capacity to train models that have a limited amount of labeled data, by utilizing additional unlabeled data that is available. We show how FL combined with AL or SSL can realize a detection system that adapts and trains quickly to new networks, reducing the total amount of data labeling needed. …

Evaluating The Chief Of Staff Of The Air Force 2016 Initiative To Revitalize The Squadron: A Thematic Content Analysis Of Appreciative Inquiry Mechanisms, John M. Huntz

Theses and Dissertations

A thorough thematic analysis and literature review were undertaken to understand better the integration of AI mechanisms within the Revitalize the Squadron initiative. To facilitate the initiative's implementation, the aim is to provide commanders with practical instances, dimensions, findings, and results. Throughout the coding process, instances of AI’s mechanisms were discovered in the literature. The link between PE and HQR boosted the overall vitality within the squadron, where vitality was determined to be the goal. AI, as a whole, was not found in the literature, but the analysis determined that the Revitalize the Squadron initiative was “Appreciative” in nature.

Analysis Of Multi-Agent Routing Solution Methodologies Exploring A Mosaic Warfare Strategy, Stephen D. Donnel

Theses and Dissertations

Recognizing that communication between assets may be possible locally but not globally (e.g., due to disruptions to a communication network), Mosaic Warfare requires the movement and operation of multiple, dispersed assets in smaller groups (i.e., tiles), within which exist hierarchical, functional relationships between assets. This research first evaluates a heuristic for an enterprise of aerial assets comprised of airborne sensors, command and control, and strike aircraft seeking to move towards and destroy stationary targets. Next, we examine routing multiple assets of different types over a network to service demands in a collaborative manner, in that, when servicing a demand, …

Improving Deep Reinforcement Learning Methodology For Autonomous Defense And Escort Of Military High-Value Assets, Joseph Liles Iv

Theses and Dissertations

This dissertation explores the application of machine learning to the control of autonomous unmanned combat aerial vehicles (AUCAVs). In particular, this research applies deep reinforcement learning methodologies to a defensive air combat scenario wherein a fleet of AUCAVs protects a military high-value asset (HVA). A collection of air battle management scenarios along with an original simulation environment and a set of designed computational experiments support the approximation of high-quality decision policies by employing Markov decision processes, approximate dynamic programming algorithms, and deep neural networks for value function approximation.

Network Vulnerability Identification For The Material Routing Problem, Carson G. Long

Theses and Dissertations

This dissertation considers the importance of identifying spatiotemporal vulnerabilities in ground distribution networks and uses operations research methods to formulate models that allow military logistic planners to implement prevention and mitigation measures regarding the routing of personnel, equipment, and supplies in contested Areas of Responsibility (AOR). For optimization models relating to identifying spatiotemporal network vulnerabilities in distribution networks, this work leverages game theory, mixed-integer programming, multi-objective optimization, and metaheuristics to inform mitigation measures for shipment routing. This research has three related components: the first component develops a multi-objective mathematical program to identify spatiotemporal vulnerabilities via myopic heuristic identification, in combination …

Test Problem Generation And Metaheuristic Selection For The Multidemand Multidimensional Knapsack Problem, Matthew E. Scherer

Theses and Dissertations

This work focuses on instance generation methods for the multi-demand multidimensional knapsack problem (MDMKP). Specifically, instance space analysis (ISA) is used to characterize the landscape of existing instances and validate the novelty of new instances generated with a novel problem generation method, the primal problem instance generator (PPIG). The instance generator is capable of producing feasible, diverse, and challenging instances by directly controlling the problem features. PPIG contributes to the previous collections of instances and is validated through instance space analysis. The research presents an in-depth empirical evaluation of existing solution procedures for the MDMKP. The portfolio of metaheuristics examined …

Advances In Quaternion-Valued Neural Networks, Jeremiah P. Bill

Theses and Dissertations

This dissertation investigates the construction, optimization, and application of quaternion neural networks (QNNs) to Department of Defense (DoD) related problem sets. QNNs are a type of neural network wherein the weights, biases, and input values are all represented as quaternion numbers. This work provides a critical evaluation of the myriad different quaternion backpropagation derivations that exist in the literature, testing the performance of each on a range of regression problem sets. The optimization dynamics of QNNs are explored, presenting visualizations of QNN loss surfaces and a novel method for assessing the “smoothness” of these loss surfaces. Finally, this dissertation presents …

Assessing And Predicting The Students’ Systems Thinking Preference: Multi-Criteria Decision Making And Machine Learning, Siham Tazzit

Theses and Dissertations

The 21st century is marked by a technological revolution that features digital implementation and high interconnectivity between systems across different domains, such as transportation, agriculture, education, and health. Although these technological changes resulted in modern systems capable of easing individuals’ lives, these systems are increasingly complex, and that increased complexity is only expected to continue. The increased system complexity is due to the rapid exchange of information between subsystems, which creates high interconnectivity and interdependence between the subsystems and their elements. Workforce skill sets, as a result, must be modified appropriately to ensure the systems’ success. Systems Thinking is an …

Ruggedness Test Of A New Standardized Test Method For Abrasion Resistance Of E-Textiles, Erin Parker

Theses and Dissertations

Standard test methods provide product developers with information regarding materials' suitability for different purposes. Typically, current standards are suitable for determining the mechanical properties of new materials. However, in the case of electronic textiles (E-Textiles) and wearable technology (wearables), adding conductive components with added functionality makes utilizing textile standards difficult, and these standards will not provide information on mechanical and electrical properties of conductive elements. New standards for E-Textile and wearables testing are needed to ensure product developers can obtain the information necessary to make informed decisions about new products. Standards organizations such as the American Society for Testing and …

Ai Methods For Identifying Process Defects In Advanced Manufacturing With Rare Labeled Data, Ayantha Umesh Senanayaka Mudiyanselage

Theses and Dissertations

This dissertation aims to provide efficient process defect identification methods for advanced manufacturing environments using AI tools/algorithms with limited labeled data availability. Asset and equipment quality become highly sensitive in sustaining virtuous performance and safety in various manufacturing domains. Internally generated process imperfections degrade finished products' optimum performance and mechanical attributes. The evolution of big data and intelligent sensing systems leverage data-driven defect identification in advanced manufacturing environments. Widely adopted data-driven process anomaly detection methods assume that the training (source) and testing (target) data follow the same distribution and that labeled data are available in both source and target domains. …