Operational Research Commons^™

Models And Algorithms For Trauma Network Design., Sagarkumar Dhirubhai Hirpara

Electronic Theses and Dissertations

Trauma continues to be the leading cause of death and disability in the US for people aged 44 and under, making it a major public health problem. The geographical maldistribution of Trauma Centers (TCs), and the resulting higher access time to the nearest TC, has been shown to impact trauma patient safety and increase disability or mortality. State governments often design a trauma network to provide prompt and definitive care to their citizens. However, this process is mainly manual and experience-based and often leads to a suboptimal network in terms of patient safety and resource utilization. This dissertation fills important …

Recall Distortion In Neural Network Pruning And The Undecayed Pruning Algorithm, Aidan Good, Jiaqi Lin, Hannah Sieg, Mikey Ferguson, Xin Yu, Shandian Zhe, Jerzy Wieczorek, Thiago Serra

Faculty Conference Papers and Presentations

Pruning techniques have been successfully used in neural networks to trade accuracy for sparsity. However, the impact of network pruning is not uniform: prior work has shown that the recall for underrepresented classes in a dataset may be more negatively affected. In this work, we study such relative distortions in recall by hypothesizing an intensification effect that is inherent to the model. Namely, that pruning makes recall relatively worse for a class with recall below accuracy and, conversely, that it makes recall relatively better for a class with recall above accuracy. In addition, we propose a new pruning algorithm aimed …

Using Strategic Options Development And Analysis (Soda) To Understand The Simulation Accessibility Problem, Andrew J. Collins, Ying Thaviphoke, Antuela A. Tako

Engineering Management & Systems Engineering Faculty Publications

Simulation modelling is applied to a wide range of problems, including defense and healthcare. However, there is a concern within the simulation community that there is a limited use and implementation of simulation studies in practice. This suggests that despite its benefits, simulation may not be reaching its potential in making a real-world impact. The main reason for this could be that simulation tools are not widely accessible in industry. In this paper, we investigate the issues that affect simulation modelling accessibility through a workshop with simulation practitioners. We use Strategic Options Development and Analysis (SODA), a problem-structuring approach that …

Study Of Stochastic Market Clearing Problems In Power Systems With High Renewable Integration, Saumya Sakitha Sashrika Ariyarathne

Operations Research and Engineering Management Theses and Dissertations

Integrating large-scale renewable energy resources into the power grid poses several operational and economic problems due to their inherently stochastic nature. The lack of predictability of renewable outputs deteriorates the power grid’s reliability. The power system operators have recognized this need to account for uncertainty in making operational decisions and forming electricity pricing. In this regard, this dissertation studies three aspects that aid large-scale renewable integration into power systems. 1. We develop a nonparametric change point-based statistical model to generate scenarios that accurately capture the renewable generation stochastic processes; 2. We design new pricing mechanisms derived from alternative stochastic programming …

Heuristics For Capacity Allocation And Queue Assignment In Congested Service Systems With Stochastic Customer Demand And Immobile Servers, Adam Colley

Operations Research and Engineering Management Theses and Dissertations

We propose easy-to-implement heuristics for a problem referred to in the literature as the facility location problem with immobile servers, stochastic demand, and congestion, or the service system design problem. The problem is typically posed as one of allocating capacity to a set of M/M/1 queues to which customers with stochastic demand are assigned with the objective of minimizing a cost function composed of a fixed capacity-acquisition cost, a variable customer-assignment cost, and an expected-waiting-time cost. The expected-waiting-time cost results in a non-linear term in the objective function of the standard binary programming formulation of the problem. Thus, the solution …

Retention Prediction And Policy Optimization For United States Air Force Personnel Management, Joseph C. Hoecherl

Theses and Dissertations

Effective personnel management policies in the United States Air Force (USAF) require methods to predict the number of personnel who will remain in the USAF as well as to replenish personnel with different skillsets over time as they depart. To improve retention predictions, we develop and test traditional random forest models and feedforward neural networks as well as partially autoregressive forms of both, outperforming the benchmark on a test dataset by 62.8% and 34.8% for the neural network and the partially autoregressive neural network, respectively. We formulate the workforce replenishment problem as a Markov decision process for active duty enlisted …

Optimizing Incentives For Systems With Heterogeneous Agents, Chen Chen

Dissertations

This dissertation explores new models and applications based on the game theory of incentives. This exploration starts with controlling an invasive insect problem to address one of the most significant challenges facing our forests, the invasion of the Emerald ash borer (EAB), a non-native, wood-boring insect that threatens to kill most ash trees in North America, through designing two new cost-sharing programs between the landowners and local governments. Ash trees are one of North America’s most widely distributed tree genera and a vital part of the green infrastructure of cities, where they provide residents with numerous social, economic, and ecological …

Developing Novel Optimization And Machine Learning Frameworks To Improve And Assess The Safety Of Workplaces, Amin Aghalari

Theses and Dissertations

This study proposes several decision-making tools utilizing optimization and machine learning frameworks to assess and improve the safety of the workplaces. The first chapter of this study presents a novel mathematical model to optimally locate a set of detectors to minimize the expected number of casualties in a given threat area. The problem is formulated as a nonlinear binary integer programming model and then solved as a linearized branch-and-bound algorithm. Several sensitivity analyses illustrate the model's robustness and draw key managerial insights. One of the prevailing threats in the last decades, Active Shooting (AS) violence, poses a serious threat to …

Optimization Of Drive Time And Competitiveness In Sports League Design, Zhuo Chen

Operations Research and Engineering Management Theses and Dissertations

Club sports, also known as recreational team sports, are prevalent in the metropolitan areas of United States nowadays. However, there is a key concern for organizers, which is how to reduce the time that players spend driving to and from matches while keeping league divisions competitive. We adopt a three-step approach to solve this problem. Initially, we analyze the drive time data between clubs’ locations to determine the geographic regions for the league. And then, clubs are assigned to divisions based on their rankings within in the league as well as their home facilities’ geographic regions. Finally, divisions are further …

Optimizing Strategic Planning With Long-Term Sequential Decision Making Under Uncertainty: A Decomposition Approach, Zeyu Liu

Doctoral Dissertations

The operations research literature has seen decision-making methods at both strategic and operational levels, where high-level strategic plans are first devised, followed by long-term policies that guide future day-to-day operations under uncertainties. Current literature studies such problems on a case-by-case basis, without a unified approach. In this study, we investigate the joint optimization of strategic and operational decisions from a methodological perspective, by proposing a generic two-stage long-term strategic stochastic decision-making (LSSD) framework, in which the first stage models strategic decisions with linear programming (LP), and the second stage models operational decisions with Markov decision processes (MDP). The joint optimization …

Optimal Global Supply Chain And Warehouse Planning Under Uncertainty, Avnish Kishor Malde

All Dissertations

A manufacturing company's inbound supply chain consists of various processes such as procurement, consolidation, and warehousing. Each of these processes is the focus of a different chapter in this dissertation.

The manufacturer depends on its suppliers to provide the raw materials and parts required to manufacture a finished product. These suppliers can be located locally or overseas with respect to the manufacturer's geographic location. The ordering and transportation lead times are shorter if the supplier is located locally. Just In Time (JIT) or Just In Sequence (JIS) inventory management methods could be practiced by the manufacturer to procure the raw …

Adaptive Design And Flexible Approval Of Clinical Trials, Saeid Delshad Sisi

All Dissertations

Dose-finding clinical trials are among the most critical cornerstones of the healthcare system. In this broad research area, there are many decision making problems that are extremely challenging to address. However, a small improvement may result in significant benefits to the society. Dose-finding clinical trials are extremely expensive and require multiple time-consuming and complicated R&D phases. Despite all the costs and the long time these trials need to conclude (on average over ten years for each new drug/technology), only less than 15\% of these trials successfully end up in a new approved drug entering the market. This problem is even …

Park Equity Modeling: A Case Study Of Asheville, North Carolina, Anisa Young

All Theses

Parks and greenspaces are publicly available entities that serve the vital purpose of promoting multiple aspects of human welfare. Unfortunately, the existence of park disparities is commonplace within the park setting. Specifically, marginalized individuals encounter limited park access, insufficient amenity provision, and poor maintenance. To remedy these disparities, we propose a process in which we select candidate park facilities and utilize facility location models to determine the optimal primary parks from both existing and candidate sites.

We note that platforms currently exist to identify the geographical areas where residents lack sufficient access to parks. However, these platforms do not yet …

Modeling The Impact And Accelerating The Process Of Transitioning To A Sustainable Healthy Diet Through Decision Support Systems, Prince Agyemang

Graduate Theses and Dissertations

Food production and consumption are essential in human existence, yet they are implicated in the high occurrences of preventable chronic diseases and environmental degradation. Although healthy food may not necessarily be sustainable and vice versa, there is an opportunity to make our food both healthy and sustainable. Attempts have been made to conceptualize how sustainable healthy food may be produced and consumed; however, available data suggest a rise in the prevalence of health-related and negative environmental consequences of our food supply. Thus, the transition from conceptual frameworks to implementing these concepts has not always been effective. This paper explores the …

Design And Operations Of A New Facility In A Next Generation Logistic System Based On Horizontal Collaboration, Dilhani Marasinghe

All Dissertations

The growing amount of freight combined with systemic inefficiencies are stressing current logistic systems. Including horizontal collaboration in the future could provide significant help but this strategy will require several new types of facilities. The most flexible type of horizontal collaboration, however, required freight routing decisions to be made dynamically, in real time, and based on last minute information. This research explores one new facility type operating in this environment that handles a high throughput of pallets or pallet-like containers. A key design feature is how much storage these facilities should have to better coordinate outbound loads. The approach taken …

Implementation Of Lean And Six Sigma Methodologies To Improve The Operations And Efficiencies Of An Inpatient Pharmacy, Brian Ricks

Medical Engineering Theses

Reducing overhead costs and eliminating process waste are important aspects of any successful organization. Hospital processes are frequently interconnected, with many exchanges of both materials and information between departments. The Inpatient Pharmacy, a key component of any hospital, has hundreds of interactions between departments on any given day, and many processes see the pharmacy acting as producers, consumers, and transporters of goods throughout the hospital. Such a varied role provides broad opportunities for process improvement. Within the broader manufacturing world, the philosophies of Lean and Six Sigma have helped many companies increase their process efficiencies, reduce costs, and increase quality. …

Credit Card Fraud Detection Using Machine Learning Techniques, Nermin Samy Elhusseny, Shimaa Mohamed Ouf, Amira M. Idrees Ami

Future Computing and Informatics Journal

This is a systematic literature review to reflect the previous studies that dealt with credit card fraud detection and highlight the different machine learning techniques to deal with this problem. Credit cards are now widely utilized daily. The globe has just begun to shift toward financial inclusion, with marginalized people being introduced to the financial sector. As a result of the high volume of e-commerce, there has been a significant increase in credit card fraud. One of the most important parts of today's banking sector is fraud detection. Fraud is one of the most serious concerns in terms of monetary …

Training Thinner And Deeper Neural Networks: Jumpstart Regularization, Carles Riera, Camilo Rey, Thiago Serra, Eloi Puertas, Oriol Pujol

Faculty Conference Papers and Presentations

Neural networks are more expressive when they have multiple layers. In turn, conventional training methods are only successful if the depth does not lead to numerical issues such as exploding or vanishing gradients, which occur less frequently when the layers are sufficiently wide. However, increasing width to attain greater depth entails the use of heavier computational resources and leads to overparameterized models. These subsequent issues have been partially addressed by model compression methods such as quantization and pruning, some of which relying on normalization-based regularization of the loss function to make the effect of most parameters negligible. In this work, …

Reconciliation, Restoration And Reconstruction Of A Conflict Ridden Country, Muhammad S. Riaz

Theses and Dissertations

Conflict has sadly been a constant part of history. Winning a conflict and making a lasting peace are often not the same thing. While a peace treaty ends a conflict and often dictates terms from the winners’ perspective, it may not create a lasting peace. Short of unconditional surrender, modern conflict ends with a negotiated cessation of hostilities. Such accords may have some initial reconstruction agreements, but Reconciliation, Restoration and Reconstruction (RRR) is a long term process. This study maintains that to achieve a lasting peace: 1) The culture and beliefs of the conflict nation must be continuously considered and …

Optimization And Information Problems In Operations, Puping Jiang

Olin Business School Electronic Theses and Dissertations

The main purpose of this dissertation is to study the optimization problems and the value of information in various commercial settings, especially in the emerging platform economy.

Chapter 1, “Data-Driven Asset Selling”. Motivated by online asset selling marketplace business (e.g., used cars and real estate), we formulate a data-driven asset selling dynamic pricing framework which utilizes platforms’ access to customers’ online behavioral data. With mild assumptions on the demand model, careful characterization of the problem structure shows that the model admits some ideal properties that facilitate our regret analysis under our dynamic programming setting. Instead of studying the policy performance …