Operations Research, Systems Engineering and Industrial Engineering Commons^™

Reliability Modeling And Improvement Of Critical Infrastructures: Theory, Simulation, And Computational Methods, José Carlos Hernández Azucena

Graduate Theses and Dissertations

This dissertation presents a framework for developing data-driven tools to model and improve the performance of Interconnected Critical Infrastructures (ICIs) in multiple contexts. The importance of ICIs for daily human activities and the large volumes of data in continuous generation in modern industries grant relevance to research efforts in this direction. Chapter 2 focuses on the impact of disruptions in Multimodal Transportation Networks, which I explored from an application perspective. The outlined research directions propose exploring the combination of simulation for decision-making with data-driven optimization paradigms to create tools that may provide stakeholders with optimal policies for a wide array …

Using Reinforcement Learning To Improve Network Reliability Through Optimal Resource Allocation, Henley Wells

Graduate Theses and Dissertations

Networks provide a variety of critical services to society (e.g. power grid, telecommunication, water, transportation) but are prone to disruption. With this motivation, we study a sequential decision problem in which an initial network is improved over time (e.g., by adding or increasing the reliability of edges) and rewards are gained over time as a function of the network’s all-terminal reliability. The actions during each time period are limited due to availability of resources such as time, money, or labor. To solve this problem, we utilized a Deep Reinforcement Learning (DRL) approach implemented within OpenAI-Gym using Stable Baselines. A Proximal …

Deep Learning Applications In Industrial And Systems Engineering, Winthrop Harvey

Graduate Theses and Dissertations

Deep learning - the use of large neural networks to perform machine learning - has transformed the world. As the capabilities of deep models continue to grow, deep learning is becoming an increasingly valuable and practical tool for industrial engineering. With its wide applicability, deep learning can be turned to many industrial engineering tasks, including optimization, heuristic search, and functional approximation. In this dissertation, the major concepts and paradigms of deep learning are reviewed, and three industrial engineering projects applying these methods are described. The first applies a deep convolutional network to the task of absolute aerial geolocalization - the …

Supervised Representation Learning For Improving Prediction Performance In Medical Decision Support Applications, Phawis Thammasorn

Graduate Theses and Dissertations

Machine learning approaches for prediction play an integral role in modern-day decision supports system. An integral part of the process is extracting interest variables or features to describe the input data. Then, the variables are utilized for training machine-learning algorithms to map from the variables to the target output. After the training, the model is validated with either validation or testing data before making predictions with a new dataset. Despite the straightforward workflow, the process relies heavily on good feature representation of data. Engineering suitable representation eases the subsequent actions and copes with many practical issues that potentially prevent the …

The Impact Of Reliability In Conceptual Design - An Integrated Trade-Off Analysis, Tevari James Barker

Graduate Theses and Dissertations

Research presented in this paper focuses on developing models to estimate the systemreliability of Unmanned Ground Vehicles using knowledge and data from similar systems. Traditional reliability approaches often require detailed knowledge of a system and are used in later design stages as well as development, operational test and evaluation, and operations. The critical role of reliability and its impact on acquisition program performance, cost, and schedule motivate the need for improved system reliability models in the early design stages. Reliability is often a stand-alone requirement and not fully included in performance and life cycle cost models. This research seeks to …

Quantitative Set-Based Design For Complex System Development, Nicholas J. Shallcross

Graduate Theses and Dissertations

This dissertation comprises a body of research facilitating decision-making and complex system development with quantitative set-based design (SBD). SBD is concurrent product development methodology, which develops and analyzes many design alternatives for longer time periods enabling design maturation and uncertainty reduction. SBD improves design space exploration, facilitating the identification of resilient and affordable systems. The literature contains numerous qualitative descriptions and quantitative methodologies describing limited aspects of the SBD process. However, there exist no methodologies enabling the quantitative management of SBD programs throughout the entire product development cycle. This research addresses this knowledge gap by developing the process framework and …

Efficiently Estimating Survival Signature And Two-Terminal Reliability Of Heterogeneous Networks Through Multi-Objective Optimization, Daniel Bruno Lopes Da Silva

Graduate Theses and Dissertations

The two-terminal reliability problem is a classical reliability problem with applications in wired and wireless communication networks, electronic circuit design, computer networks, and electrical power distribution, among other systems. However, the two-terminal reliability problem is among the hardest combinatorial problems and is intractable for large, complex networks. Several exact methods to solve the two-terminal reliability problem have been proposed since the 1960s, but they have exponential time complexity in general. Hence, practical studies involving large network-type systems resort to approximation methods to estimate the system's reliability. One attractive approach for quantifying the reliability of complex systems is to use signatures, …

Towards A Cyber-Physical Manufacturing Cloud Through Operable Digital Twins And Virtual Production Lines, Md Rakib Shahriar

Graduate Theses and Dissertations

In last decade, the paradigm of Cyber-Physical Systems (CPS) has integrated industrial manufacturing systems with Cloud Computing technologies for Cloud Manufacturing. Up to 2015, there were many CPS-based manufacturing systems that collected real-time machining data to perform remote monitoring, prognostics and health management, and predictive maintenance. However, these CPS-integrated and network ready machines were not directly connected to the elements of Cloud Manufacturing and required human-in-the-loop. Addressing this gap, we introduced a new paradigm of Cyber-Physical Manufacturing Cloud (CPMC) that bridges a gap between physical machines and virtual space in 2017. CPMC virtualizes machine tools in cloud through web services …

Improvement Of Geometric Quality Inspection And Process Efficiency In Additive Manufacturing, Yu Jin

Graduate Theses and Dissertations

Additive manufacturing (AM) has been known for its ability of producing complex geometries in flexible production environments. In recent decades, it has attracted increasing attention and interest of different industrial sectors. However, there are still some technical challenges hindering the wide application of AM. One major barrier is the limited dimensional accuracy of AM produced parts, especially for industrial sectors such as aerospace and biomedical engineering, where high geometric accuracy is required. Nevertheless, traditional quality inspection techniques might not perform well due to the complexity and flexibility of AM fabricated parts. Another issue, which is brought up from the growing …

Low Latency Anomaly Detection With Imperfect Models, Samrat Nath

Graduate Theses and Dissertations

The problem of anomaly detection deals with detecting abrupt changes/anomalies in the distribution of sequentially observed data in a stochastic system. This problem applies to many applications, such as signal processing, intrusion detection, quality control, medical diagnosis, etc. A low latency anomaly detection algorithm, which is based on the framework of quickest change detection (QCD), aims at minimizing the detection delay of anomalies in the sequentially observed data while ensuring satisfactory detection accuracy. Moreover, in many practical applications, complete knowledge of the post-change distribution model might not be available due to the unexpected nature of the change. Hence, the objective …

Convergent Set-Based Design In Integrated Analysis Of Alternatives: Designing Engineered Resilient Systems, Zephan Wright Wade

Graduate Theses and Dissertations

This thesis presents a comprehensive package for understanding and expanding set-based design quantification through the definition and demonstration of Convergent set-based design (SBD). Convergent SBD is a technique developed for the Engineered Resilient Systems program sponsored by the Department of Defense. Convergent SBD contributes a repeatable methodology with the goal of mathematically eliminating inefficient sets. The study of Convergent SBD led to the development of dominance identification criteria equations using comparison of statistical means. The demonstration of Convergent SBD also illustrates the effect of mission resilience in the tradespace and the impact mission resilience has on preference. Finally, Convergent SBD …

Demonstrating Set-Based Design Techniques- A Uav Case Study, Colin Small

Graduate Theses and Dissertations

The Department of Defense (DoD) and Engineered Resilient Systems (ERS) community seek to improve decision making in the Analysis of Alternatives (AoA) process by incorporating resilience and leveraging the capabilities of model-based engineering (MBE) early in the design process. Traditional tradespace exploration utilizing Point-Based Design (PBD) often converges quickly on a solution with subsequent engineering changes to modify the design. However, this process can lead to a suboptimal solution if an incorrect initial solution is chosen. Enabled by MBE, Set-Based Design (SBD) considers sets of all possible solutions and enables down-selecting possibilities to converge on a final solution. Using a …

Grouping Techniques To Manage Large-Scale Multi-Item Multi-Echelon Inventory Systems, Anvar Abaydulla

Graduate Theses and Dissertations

Large retail companies operate large-scale systems which may consist of thousands of stores. These retail stores and their suppliers, such as warehouses and manufacturers, form a large-scale multi-item multi-echelon inventory supply network. Operations of this kind of inventory system require a large number of human resources, computing capacity, etc.

In this research, three kinds of grouping techniques are investigated to make the large-scale inventory system “easier” to manage. The first grouping technique is a network based ABC classification method. A new classification criterion is developed so that the inventory network characteristics are included in the classification process, and this criterion …

Essays On Optimization And Modeling Methods For Reliability And Reliability Growth, Thomas Paul Talafuse

Graduate Theses and Dissertations

This research proposes novel solution techniques in the realm of reliability and reliability growth. We first consider a redundancy allocation problem to design a system that maximizes the reliability of a complex series-parallel system comprised of components with deterministic reliability. We propose a new meta-heuristic, inspired by the behavior of bats hunting prey, to find component allocation and redundancy levels that provide optimal or near-optimal system reliability levels. Each component alternative has an associated cost and weight and the system is constrained by cost and weight factors. We allow for component mixing within a subsystem, with a pre-defined maximum level …

Surveillance Planning Against Smart Insurgents In Complex Terrain, Nabil Lehlou

Graduate Theses and Dissertations

This study is concerned with finding a way to solve a surveillance system allocation problem based on the need to consider intelligent insurgency that takes place in a complex geographical environment. Although this effort can be generalized to other situations, it is particularly geared towards protecting military outposts in foreign lands. The technological assets that are assumed available include stare-devices, such as tower-cameras and aerostats, as well as manned and unmanned aerial systems. Since acquiring these assets depends on the ability to control and monitor them on the target terrain, their operations on the geo-location of interest ought to be …