Operations Research, Systems Engineering and Industrial Engineering Commons^™

Crime Theory Informed Agent-Based Modeling For Crime Prediction And Patrolling Route Optimization, Shohreh Moradi

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Crime reduction remains a global priority, demanding both accurate modeling of criminal dynamics and efficient allocation of scarce policing resources. To address these needs, this study presents a two‐fold framework that (1) simulates street‐level crime patterns using an agent‐based model (ABM) grounded in Routine Activity Theory (RAT), Rational Choice Theory (RCT), and Crime Pattern Theory (CPT), and (2) optimizes patrol routing through a time-dependent, multi‐visit mixed‐integer linear programming (MILP) formulation.

In the first component, we integrate real‐world crime, environmental, and census data to reproduce realistic offender, citizen, and Police behaviors, capturing where and when robbery, burglary, and larceny occur across …

Enhanced Load Detection With Data-Driven Appliance Signatures Using Mixed Integer Linear Programming In Non-Intrusive Load Monitoring, Marina Materikina

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Despite the numerous research studies and interest in the non-intrusive load monitoring (NILM) area to improve energy efficiency, the problem of accurate and precise disaggregation of electrical devices has not been solved yet. The goal of our research is to build a method with a focus on higher accuracy on complex state-based appliances, which most approaches struggle to detect due to their power signal complexity and low consumption. Our approach is NILM with data-driven signatures (DS), with the ability to potentially predict power usage over time that would work great for suitable applications such as demand response, anomaly detection, and …

A Design And Analysis Of Computer Experiments Approach To Water Distribution Network Seismic Rehabilitation Optimization, Uthman Abiola Kareem

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Water is an essential part of human life. However, there are critical infrastructures that enable water availability in communities and homes. One of such is a water distribution network. Water distribution network performance depends on its reliability, which could be threatened by external agents like earthquakes. When earthquakes occur, they cause damages on some pipes within the distribution network and this limits performance of water distribution network. While earthquakes cannot be prevented, effective maintenance intervention may reduce the impact of earthquakes on water distribution networks. In order to develop an effective maintenance plan, researchers approach it in different ways. However, …

Analysis Of Crowd Logistics Networks Using Agent-Based Models, Preetam Kulkarni

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Crowd logistics is a system in which an online platform connects a group of non-professional couriers (crowd/carriers), who use their under-utilized resources to offer delivery service to other individuals or businesses (senders) for a fee. While crowd logistics platforms have the potential to offer more flexible and responsive delivery services for much lower rates than traditional logistics providers, it is difficult for platforms to be successful as it is challenging to meet carriers’ and senders’ expectations. Crowd logistics has been applied in the context of food and grocery delivery, parcel pickup and drop-off services and last-mile delivery, however, it has …

Representation Learning Of Point Cloud Data For Process Mining And Anomaly Detection In Complex Systems, Yujing Yang

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Complex systems, e.g., advanced manufacturing systems, are largely associated with dynamic and transient behaviors, resulting in condition changes and anomalies. Sensor-based condition monitoring is critical in detecting anomalies and supporting process monitoring and performance improvement for complex manufacturing systems. Traditional sensor-based monitoring approaches primarily focus on one-dimensional (1D) signals and two-dimensional (2D) images, which are limited in their ability to capture high-resolution spatial patterns pertaining to anomalies induced by systems’ condition changes, especially subtle ones. Recent advancements in three-dimensional (3D) sensing present a unique opportunity to address this limitation by enabling the capture of 3D point cloud data with micro-level …

3d Point Cloud Sensing And Analytics With Applications In Process Mining And Quality Control Of Additive Manufacturing, Zehao Ye

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

The rapid advancement in Additive Manufacturing (AM) technologies has developed significant innovations in various sectors, including medical, aerospace, and automotive industries. Despite these benefits, the adoption of AM is often hindered by quality inconsistencies related to the surface defects and geometrical inaccuracies in the fabricated products. These defects can significantly undermine the mechanical properties of the products, leading to material waste and potential safety issues. This dissertation addresses the critical challenges in quality control of AM processes through the integration of 3D point cloud data and machine learning techniques, aiming to enhance the reliability and efficiency of AM systems. The …

Probabilistic Multivariate Time Series Forecasting And Robust Uncertainty Quantification With Applications In Electricity Price Prediction, Jie Han

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Electricity price forecasting (EPF) is a crucial task for market participants seeking informed decisions in day-ahead electricity markets. The increasing penetration of stochastic renewable energy and the deregulation of electricity markets pose challenges to electricity price forecasting. Given the dependence of electricity prices on stochastic factors such as weather conditions, market dynamics, and customer behaviors, deterministic forecasting methods offer limited insight into the potential future states of energy prices in highly stochastic markets. In this study, a transformer-based electricity price forecasting (TDEPF) model was developed, utilizing a two-step training process and demonstrating superior performance compared to typical RNN models. Subsequently, …

Topics In Optimization For Sustainable Energy Planning, Bahareh Nasirian

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

ABSTRACT: Due to the current trend of rising energy demand, finding alternate energy sources is vital. Many cities consider renewable energy as a component of a sustainable future. Typically, organic wastes are considered renewable energy sources. Wastes can be converted to proper energy forms using waste-to-energy technologies. On the other hand, more renewable energy sources in the power system may increase energy market stochasticity, alter system operation, and pose new problems for the current supply and demand equilibrium, which consequently requires new control methods. Hence, taking into account the issues mentioned above, in this research, we address organic waste conversion …

Increasing The Safety Of Bicyclists Using A Cyclist Behavior Questionnaire And A Smartphone Based Application, Anika Jannat Rimu

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

ABSTRACT: Bicycling is beneficial for health, the environment, road users’ flexibility, and personal expenses. Compared to motor vehicles, they are an active mode of transport, cause minimum pollution, are affordable, and can easily navigate through the increasing traffic all over the world. This increase in traffic, however, also increases the possibility of crashes with motor vehicles. Bicyclists, being more exposed to traffic than drivers, suffer fatal consequences from a crash. Therefore, a standard tool is required to understand bicyclist behavior on the road. This tool can provide insights into bicyclists’ behavior so that appropriate infrastructure or policy changes can be …

Towards Sustainable Additive Manufacturing: Assessment Of Cost, Greenhouse Gas Emission, And Recyclability, Lei Di

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Additive Manufacturing technologies fabricate 3D objects layer by layer following a predesigned CAD model. Owing to the unique layer-wise production method, additive manufacturing offers competitive advantages in comparison with traditional subtractive manufacturing, such as shortened production time, increased design freedom, improved manufacturing capability and complexity, and reduced manufacturing waste. Numerous research studies have been conducted to design, understand, and improve additive manufacturing technologies in order to facilitate the implementation in the supply chain. On the other hand, with the rapid growth of additive manufacturing, sustainability issues that exist on both process level and supply chain level have started to receive …

Applications Of Probability Of Success In The Well Delivery Process To Improve Risk, Opportunity, And Cost Assurance, Romar Alexandra Gonzalez Luis

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Oil and Gas (O&G) well drilling is risky and expensive. The cost of drilling is typically underestimated, and there is little understanding of the certainty or Probability of Success (POS) of achieving the well objectives within the estimated cost. This dissertation presents, for the first time, a publicly available POS Cost method/tool that enables O&G operators to estimate the cost of well drilling substantially more accurately and improve the POS of accomplishing the drilling for the estimated cost. This POS Cost method employs a comprehensive, expert-based assessment of risks and risk mitigations that are incorporated into a Monte-Carlo-based simulation of …

Machine Learning For Ultraviolet Spectral Prediction, Linh Ho Manh

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Machine Learning has found wide applications in material science, including dielectric polymers, superconducting materials, and drug property prediction. The use of data analytics and machine learning methods to predict Vacuum Ultraviolet (VUV) spectra by encoding molecular structure is gaining interest because high-quality VUV spectral prediction capability would enable the study of new molecules without costly wet-lab measurements. This dissertation aims to study feature representations for molecular structures that enhance the prediction of VUV spectra via machine learning models. Both interpretable machine learning and deep learning are studied. Chapter 1 provides an overview of VUV/UV spectra retrieval, and Chapter 2 reviews …

Forming Coalitions And Sharing Payoffs In N-Person Normal Form Games, Emma Owusu Dwobeng

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

For a given n-person normal form game, we form all possible sets of mutually exclusive and collectively exhaustive coalitions of the n players. For each set of coalitions, we define a coalitional semi-cooperative game as one in which these coalitions are taken as the players of this new game, each coalition tries to maximize the sum of its individual players’ payoffs, and the players within a coalition cooperate to do so. For any coalitional semi-cooperative game, the goal of the original n players is to improve their individual payoffs obtained in a Greedy Scalar Equilibrium (GSE) of the original game, …

Lasso Based State Transition Modeling With Interactions In Adaptive Interdisciplinary Pain Management, Amith Viswanatha

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

The Eugene McDermott Center for Pain Management at the University of Texas Southwestern Medical Center has an interdisciplinary pain management program for chronic pain. This program treats patients with a holistic view of reducing chronic pain and improving their physical, mental, and social well-being through treatment interventions. The development of an adaptive treatment decision tool is main goal of the research project. This program is modeled as a two-stage adaptive treatment decision problem, with state transition models representing the transition of patient state, treatment, and outcome variables from stage 1 to stage 2. Interactions between the patient state and treatments …

A Hybrid Systems Model For Emergency Department Boarding Management, Eniola Oluwasola Suley

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

The purpose of this research is to examine methods for minimizing the influence of boarding on emergency department (ED) crowding outcomes. To accomplish this purpose, this research uses a hybrid systems model framework by combining agent-based simulation, predictive and optimization models to improve ED outcomes such as length-of-stay and left-without-being-seen rates. For the research, different types of simulation models were examined (discrete event and agent-based/discrete event combination) to identify the most parsimonious for studying ED boarding. Predictive models using simulation output were developed to understand the factors that influence future boarding levels as well as generate predictions. Research has previously …

Optimizing The Performance Of Analytical Chemistry Instrumentation, Srividya Sekar

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Surrogate Optimization and Global Optimization approaches to optimize underlying functions have been studied and used extensively in the field of Operations Research. However, there are very few instances where these approaches have been applied and tested in applications with uncertainty. Additionally, extensive focus and effort have been put into developing highly complex metamodels rather than globally optimizing these metamodels. In this study, we propose a Mixed Integer Quadratically Constrained Program (MIQCP) based approach that globally optimizes a Quintic Multivariate Adaptive Regression Splines (QMARS) metamodel. The QMARS-MIQCP based optimization is applied to a global optimization framework called QMARS-MIQCP-OPT to optimize several …

Evaluating And Addressing The Transportation Challenges Of Small-Scale Farmers And Ranchers In Regional Food Systems, Narjes Sadeghiamirshahidi

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

During the last 20 years, there has been a significant increase in US consumers’ interest in local food. At the same time, there is substantial potential demand for crops produced by local small and mid-size farmers and ranchers whose market channels are short distance distributions Although there is a variety of definitions for local food, it often refers to direct-to-customer market channels for farmers and ranchers, including farmers’ markets and Community Supported Agriculture (CSA), as well as distribution through local businesses like restaurants, grocery stores, as well as local institutions such as hospitals and schools. However, small and mid-size farmers …

Leveraging Ai And Supply Chain Technologies With Thermal Imaging And Telemedicine For Early Detection And Prevention Of Covid-19 And Respiratory Infections In Urm Communities, Gohar Azeem

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

The underserved population could be at risk during the times of crisis, unless there is strong involvement from government agencies such as local and state Health departments and federal Center for Disease Control (CDC). The COVID-19 pandemic was a crisis of different proportion, creating a different type of burden on government agencies. Vulnerable communities including the elderly populations and communities of color have been especially hard hit by this pandemic. This forced these agencies to change their strategies and supply chains to support all populations receiving therapeutics. The National Science Foundation (NSF Award # 2028612) funded this research to help …

Affordable Autonomous Vehicles For Deployment After Disastrous Events, Shannon Abolmaali

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

In disastrous events such as hurricanes and tornadoes, it has been observed that people get stranded and helpless without a feasible way to escape during those emergency situations. This became very evident during hurricanes, such as Katrina and Ida affecting millions of people seeking immediate rescue efforts. With the use of artificial intelligence and machine learning, we envision an autonomous vehicle, AV, which is able to find the most optimal and safest way to help those who are stranded to get them to a safe location. Electric vehicles, EV, and Autonomous Vehicles, AV, is becoming the future; minimizing the carbon …

Design For Older Adults – Functional Limitations And Human Factors Engineering, Megumi Sato Hice

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Background: Throughout the aging process, people experience changes in functional ability and such changes can happen in physical and mental functions. As a result, older adults develop limitations in their capabilities to perform daily activities. Although many studies have done to identify functional limitations for elderly, many older adults still face negative consequences in daily living conditions due to their functional declines. One of the possible causes may be a lack of feedback or participations of elderly users in the design process. Other reason could be a lack of consideration to identify the needs for a specific task associated with …

Constraint Optimal Selection Techniques (Costs) For A Class Of Linear Programming Problems, Tai-Kuan Sung

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

This dissertation describes two classes of Constraint Optimal Selection Techniques (COSTs). An algorithm of each type is developed for solving nonnegative linear programming problems. In addition, geometric interpretations of these new algorithms are given, computational results for some large-scale problems are provided, and directions for future research are discussed.

A Systematic Stakeholder-Driven Framework For Empirical Characterization And Parameterization Of Human Agents For Agent-Based Modeling Of Older Adults' Transportation, Nilufer Oran Gibson

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Standard of living for health and well-being of oneself and of one’s family is a human right. Vulnerable populations are incapable of maintaining an adequate standard of living due to several reasons, such as financial constraints, racial profiling, health conditions, aging, and the combination of these reasons. To alleviate their vulnerability, the most basic needs of the vulnerable populations must be met. Transportation is the essential link to the resources to address the basic needs of the vulnerable populations. Transportation solutions tailored for vulnerable populations to meet their basic needs have multiple dimensions that require the involvement of all government …

Machine Learning Framework For Nonlinear And Interaction Relationships Involving Categorical And Numerical Features, Shirish Mohan Rao

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Traditionally, physical scientific experiments have been conducted extensively to study and understand the behavior of a process or a system. With the advancement of computing technology in recent years, computer codes and algorithms are used as simulators to replicate behavior of a complex system. Such use of computers to study a system is termed as ‘computer experiments.’ The process involves selecting specific points or runs in the design space in order to maximize information about the system in minimal runs. These computer models are high dimensional and can take a long time to simulate. Metamodels (or surrogate models) built using …

Agent-Based Model Simulation For Police Deployment Decision-Making In Patrol Operations, Yasaman Ghasemi

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Police patrolling plays a key role in responding to 911 calls and reducing crimes. The effectiveness of patrol operations heavily depends on the deployment of police officers – e.g., the number of officers assigned to specific policing districts or beats. The complex nature of the policing system – dynamic and stochastic criminal behavior, compounded with limited policing resources, render current (traditional) police operations, which are often managed in a reactive and stationary manner – often makes it very challenging to manage and control. This study develops an agent-based simulation framework to address the dynamically changing environment in police operations and …

Shape-Based Time Series Mining For Process Monitoring And Anomaly Detection, Li Zhang

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Due to the rapid development of computing and sensing technology, Internet of Things (IoT)-enabled monitoring plays a crucial role for people suffering from cardiac problems. It is important to detect the abnormal ECG cycles during the cardiac monitoring for the early treatment. However, most existing methods focused on the full reading of time series, for the cycle-based time series, it is wasting time to read the whole time series while we can find the characteristic patterns instead. Characteristic patterns named shapelets are time series subsequences, which are explainable and discriminative features that can best classify time series. Shapelet-based classification that …

Using Empirical Data To Design And Validate Hybrid Simulation Models Of Human Behavior In Service Operations, Mohammed Farhan

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

The purpose of this research is to examine how a change in a team member's role, team dynamics, and organizational policies impact an individual's motivation to engage in helping behavior, as well as the impact of helping behavior on service system operational performance. To analyze these behavioral dynamics in a dynamic setting, this research integrates empirical human behavioral data into a hybrid discrete-event and agent-based simulation model of service operations in a restaurant. The model was then validated using Metamorphic Testing (MT), an approach that has previously been used for verification of software. Recent research shows that MT can be …

Teenage Cyclists’ Perception Towards Autonomous Vehicles And Its Associated Traffic Infrastructures, Obiageli Lawrentia Ngwu

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Background: Cycling is a cost-effective means of transportation. Many teenagers cycle to go to schools and ride in neighborhoods. Cyclists are more vulnerable to injuries and fatalities than motor vehicle drivers. With the implementation of autonomous vehicles (AVs), interactions between AVs and road-users are expected to be safer. It is most likely that current young people will be the ones to use these vehicles and interact with them. However, very few past studies have focused on cyclist-AV interaction, with little to no attention toward the teenage cyclist population. Objectives: This study is aimed at examining teenage cyclists’ perceptions of AVs …

Bidding Enabled Inventory Redistribution In A Retail Network, Hafsa Binte Mohsin

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

This research aspires to develop a systematic approach to minimize the demand and supply gap of products and product expiration in a feasible way. Inventory replenishment policy, uncertain customer demand, and forecast inaccuracy are some of the reasons that create imbalanced stocks in the outlets. Lateral transshipment or redistribution, donation, and promotion have been discussed in the existing literature separately as ways to balance and utilize inventory. Redistribution needs to account for extra transportation costs due to stock transfer. Existing literature on redistribution fails to address products’ physical attributes, valuation of products as a function of time, and constraint on …

A Biomechanical Approach To Investigate The Effects On The Lumbosacral Joint, Pelvis, And Knee Joint While Of Carrying Asymmetrical Loads, During Ground Walking., Tomal Das

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Spinal pain is reasonably considered among the expensive and impairing problems critically disturbing the health of individuals, especially the workforce, in industrially developed countries (Steele et al., 2003). The pain adversely affecting the lumbar region or pelvis is typically considered as low back pain. As per the National Institute of Neurological Disorder and Stroke, around 80 percent of grown-ups encounter spinal pain eventually in the course of their lifetime. The possible risk factors include age, fitness level, genetics, weight gain, occupational risk factors such as having a job that expects someone to do lifting, carrying, pushing, pulling, distorting the spinal …

A Two-Stage Stochastic Programming Model For Enhancing Seismic Resilience Of Water Pipe Networks, Azam Boskabadi

Industrial, Manufacturing, and Systems Engineering Dissertations - Archive

Earthquakes are sudden and inevitable disasters that can cause enormous losses and suffering, and having accessible water is critically important for earthquake victims. To address this challenge, utility managers do preventive procedures on water pipes periodically to withstand future earthquake damage. The existing seismic vulnerability models usually consider simple methods to find the pipes to rehabilitate with highest priority. In this research, we develop an optimization approach to determine which water pipes to rehabilitate subject to a limited budget to achieve highest network serviceability after a disaster. We propose a two-stage stochastic mixed integer nonlinear program (MINLP). The MINLP model …