2012 Student Poster Presentations

Special Acknowledgement to the National Science Foundation (CMMI-1130507)

Cocircuits of Linear Matroids

John Arellano

Computational and Applied Mathematics

Rice University

I present a set covering problem (SCP) formulation of the matroid cogirth problem. Addressing the matroid cogirth problem can lead to significantly enhancing the design process of sensor networks. The solution to the matroid cogirth problem provides the degree of redundancy of the corresponding sensor network, and allows for the evaluation of the quality of the network. I provide computational results to validate a branch-and-cut algorithm that addresses the SCP formulation.

A Simulation-Optimization Approach for Large-Scale Wildfire Response Planning

Michelle Alvarado

Industrial and Systems Engineering

Texas A&M University

Extended wildfire response planning under uncertain weather conditions is challenging. We introduce a new simulation-optimization approach that integrates a fire behavior and suppression simulation model (DEVS-FIRE) and a stochastic integer programming model (SIP). Using weather forecasts, DEVS-FIRE makes predictions of fire growth and the effects of suppression efforts on the fire, and feeds this scenario data to SIP to optimize the dispatching of firefighting resources to the fire over time.

An Approach to Approximating Contributions Received through Food Bank Collections at Grocery Stores

Luther G. Brock III

Industrial and Systems Engineering

North Carolina A&T State University

This research addresses the need to approximate how much food is received by food banks through collections at grocery store. Such forecasts are the basis for more cost-effective vehicle scheduling. A feed-forward artificial neural network (FF-ANN) is used to estimate the amounts of different in-kind goods received through isolated store collections. The FF-ANN is trained using donation records provided by a food bank in the southeastern United States and compared with linear regression.

A Decision Support Framework for Healthcare Transition Programs to Reduce Hospital Readmissions

Sabrina Casucci

Industrial and Systems Engineering

University of Buffalo

High hospital readmissions signal poor quality of post-hospital care and unnecessarily increase healthcare costs. Intervention programs aim to reduce readmission rates; yet there is little support for understanding the effect of physician and patient care decisions on the program outcomes. This paper focuses on the complexities of transitional care and the design of informative stochastic models that can provide decision support to all the stakeholders.

A Joint Surveillance and Patrol Problem for Law Enforcement

Belleh Fontem

Information Systems, Statistics, and Management Science

University of Alabama

We investigate the problem of using unmanned aerial systems (UASs) as visual aids to ground agents while assigning them dangerous incidents in order to maximize the cumulative harm averted to society. We introduce a tight formulation of the resulting team orienteering problem and draw insights from theoretical analyses and numerical experiments.

Integrating Univariate Control Charts and Mahalanobis Distance for Near Zero Type II Error Monitoring

Weihong Guo

Industrial and Operations Engineering

University of Michigan

It is highly needed to ensure high product quality through online sensing monitoring for emerging manufacturing processes especially in producing mission- critical parts such as rechargeable batteries for electrical vehicles. This research proposed an integrated monitoring scheme that targets at a near zero Type II error rate by integrating univariate control charts and Mahalanobis distance. The detailed algorithms and its effective and implementation in a real-world example will be presented.

Two-Stage Stochastic Programming Model for Phlebotomist Scheduling in Hospital Laboratories

Laquanda Leaven

Industrial and Systems Engineering

North Carolina A&T State University

Laboratory medicine is vital to the healthcare system. A Two-Stage Stochastic Programming Model has been formulated to determine better phlebotomist schedules for the preanalytical stage of the testing process in hospital laboratories. The objective is to balance the workload among shifts and the workload among phlebotomists in each shift, which result in cost reductions.

Evaluation of Breast Cancer Mammography Screening Policies Considering Adherence Behavior

Mahboubeh Madadi

Industrial Engineering

University of Arkansas

The efficacy of mammography screening guidelines is highly associated with women's compliance with these recommendations. However, none of the existing policies take women's adherence behavior into consideration. Instead, perfect adherence is often assumed. In this study, a partially observable Markov model is proposed to evaluate and compare various screening policies, while incorporating variation in women's adherence behavior.

Time Varying Queues with Abandonment: A Laguerre Polynomial Approach

Jamol Pender

Operations Research and Financial Engineering

Princeton University

A time varying multi-server queueing model with abandonment is the ubiquitous choice for modeling service systems. As with many non-stationary models, we must settle for only understanding their moment behavior since other information is either very difficult to acquire or intractable. One way to analyze the moment behavior is through asymptotic methods. However, in this paper, we take a novel approach that uses the Kolmogorov forward equations. To analyze the moments of our queueing system, we expand the queueing process via Laguerre polynomials and use a truncated expansion to derive approximate expectations and covariance terms. The Laguerre polynomials are useful because they serve as a basis for the Hilbert space L2[[0,∞), ν], where ν is the exponential measure. This expansion is in contrast to the expansion of Hermite polynomials used by Massey and Pender, which was motivated by the Gaussian fluid and diffusion limits. Since the Laguerre are orthogonal with respect to the exponential measure, our approximation might be more useful for the single server setting where the Hermite expansion breaks down and the stationary distribution is exponentially distributed. We show that three terms of the polynomial series is enough to capture the most meaningful cumulant moments of the queueing process such as the mean, variance, and skewness.

Vaccine Prioritization Using Operations Research

Tim Schmoke

Industrial and Systems Engineering

Rochester Institute of Technology

A more effective vaccine prioritization process is essential to prevent the many issues that currently weaken the global vaccine supply chain. This research project aims to create a decision-support tool for prioritizing vaccine initiatives using mathematical optimization models. The result will be a tool that researchers and funding agencies can use to determine which initiatives are more effective and meet the needs of multiple target populations

A Model for Healthcare Supply Chain Coordination and Treatment Protocol Design Solutions

Lakausha Simpson

Industrial and Systems Engineering

North Carolina A&T State University

A Game Theoretic Model of Financial Contagion and Global Economic Risks

Jonathan W. Welburn

Industrial and Systems Engineering

University of Wisconsin

Global financial crises have revealed the systemic risk posed by economic contagion. We formulate a game between countries, central banks, banks, customers, and financial inter-governmental organizations to model the dynamics of borrowers and lenders. We model strategic choices, determine equilibrium solutions, and simulate the impacts of random shocks. Our conclusions enhance the understanding of global economic risk.

Analysis of Food Import Refusals as a Key Indicator of Risk for Food Safety

Jonathan W. Welburn

Industrial and Systems Engineering

University of Wisconsin

There is rising concern for risks associated with imported foods. We use import violations data from the FDA Operational and Administrative System for Import Support (OASIS) to address these concerns, quantify risks by product and country of origin, and explore the usefulness of OASIS data.

The Minimal k-core Problem for Modeling k-assemblies

Cynthia Woods

Computational & Applied Mathematics

Rice University

I present a recursive algorithm to find all minimal k-cores of a given undirected graph. This method is a modification of the Bron and Kerbosch algorithm for finding all cliques. The presented problem has applications in the area of neuroscience. For example, in the study of associative memory, a cell assembly is a group of neurons that are strongly connected and represent a "concept" of our knowledge. A k- assembly is a particular type of cell assembly. It is defined mathematically as the closure of a minimal k-core. The proposed method puts us a step closer to determining whether there exists a realistic number of k-assemblies in a given network. I provide computational and theoretical results to validate the proposed algorithm.

Optimal Control of an Emergency Room Triage and Treatment Process

Gabriel Zayas-Caban

Center for Applied Mathematics

Cornell University

Patient care oftentimes consists of assessment and treatment. These two phases are sometimes carried out by the same medical provider, so there is a question of how to prioritize the work in order to balance initial delays for care with the need to discharge patients in a timely fashion. We model a hospital emergency room triage and treatment process as a tandem queue with a single server, explore alternative service disciplines under various scenarios, and identify optimal policies for each.

A Decision Support Framework for Healthcare Transition Programs to Reduce Hospital Readmissions

Yuanhui Zhang

Edward P. Fitts Department of Industrial and Systems Engineering

North Carolina State University

The main objective for care of type 2 diabetes is to control the patient's glycated hemoglobin (HbA1c) to reduce the risk of the diabetes complications. Uncertainty in the progression of HbA1c and the treatment effects make treatment decisions challenging. We present a Markov decision process to maximize the patient's expected quality-adjusted life years prior to major complications. We present the structure and influence factors of the optimal policy and compare it to current guidelines.