Informs 2015

79 / 552

INFORMS Philadelphia – 2015
77
SB35
35-Room 412, Marriott
Joint Session HAS/Analytics: Predictive Models for
Clinical and Public Health Decision Making
Sponsor: Health Applications
Sponsored Session
Chair: Ozgur Araz, University of Nebraska Lincoln,
College of Business Administration, Lincoln, United States of America,
1 - Risk Factors for Disease Progression in Sepsis Patients:
A Retrospective Cohort Study
Benjamin Whitsitt, University of Nebraska Medical Center,
42nd and Emile, Omaha, NE, 68198, United States of America,
, Micah Beachy, Lorena Baccaglini,
Gleb Haynatzki, Michael Ash, Ozgur Araz
In this study we identify risk factors associated with the progression of sepsis to
severe sepsis and/or septic shock and estimate the likelihood of disease
progression for different patient groups. We also assess the probability of death
and readmission amongst a sepsis population at two area hospitals in Omaha,
Nebraska. Multiple logistic regression model is used to determine the likelihood of
disease progression, mortality, and readmission.
2 - The Impact of Geographic Localization of Patients on
Hospital Performance
Paul Cronin, PhD Student, University of Texas at Austin, 2110
Speedway Stop B6500, Austin, TX, 78712, United States of
America, , Douglas Morrice,
Jonathan Bard, Luci Leykum
We study the impact of geographic localization of patients on hospital
performance using patient-level data from a Texas teaching hospital. Performance
is measured by length of stay in ED and hospital, waiting time for bed
assignments, patient transfers between teams, and hour of discharge. The results
of this study inform admission decision-making including patient allocation to
medical teams and admission capacity planning.
3 - A Dynamic Model for Population Screening: Risk Perception and
Feedback in Screening Decision
Ozge Karanfil, PhD Candidate, MIT Sloan School of Management,
100 Main Street, E62-379, Cambridge, MA, 02142,
United States of America, John D. Sterman
In this study we built a behaviorally realistic, bounded simulation model to
explain changes in policy action thresholds of clinical practice guidelines, and to
document evidence of gaps between scientific evidence-based guidelines and
actual practice. This is the first theory building piece for cancer screening
dynamics in the US that takes into account the broader cognitive and socio-
political environment in which screening decisions are embedded.
4 - Modeling the Impact of Chronic Disease Combinations on 30-day
Hospital Readmissions
Sabrina Casucci, University at Buffalo (SUNY),
339B Bell Hall, Amherst, NY, 14260, United States of America,
Alexander Nikolaev, Li Lin, Sharon Hewner
Individuals with chronic disease are at high risk for hospital readmission. There is
significant opportunity to reduce utilization and costs by developing new
readmission reducing interventions for this population. This work reveals and
quantifies the causal impacts of specific chronic disease combinations on hospital
readmissions. The results are reported for a subset of the New York State adult
Medicaid population using observational causal inference methods and tools.
SB36
36-Room 413, Marriott
Humanitarian Applications II
Sponsor: Public Sector OR
Sponsored Session
Chair: Mahyar Eftekhar, Arizona State University, P.O. Box 874706,
Tempe, AZ, 85287, United States of America,
1 - Blood Storage and Transportation: An Important Component of
Humanitarian Logistics
Divya Nagilla, Faculty Asociate, Institute of Management
Technology, Survey No.38, Cherlaguda Village, Shamshabad, RR
District, Hyderabad, 501218, India,
Sourabh Bhattacharya
Supply of blood during and after disasters is a major part in humanitarian
logistics.The major problem during disasters is not the lack of blood supply but
disruption of the blood distribution system .A large number of practical and
logistical issues related to communication, transportation, managing donors and
volunteers arise .Depending upon the phase of the disaster the blood products
required to treat the victims vary. Safety of the blood supply and adherence to
regulations is crucial.
2 - Humanitarian Logistics in the Philippines: Case of
Typhoon Haiyan
Brian Gozun, La Salle - Universitat Ramon Lull, Carrer de Sant
Joan de La Salle 42, Barcelona, Spain,
Francesc Miralles
The study applied a humanitarian logistics framework through a deductive
approach on the experiences of various stakeholders in the Philippines after the
onslaught of Super Typhoon Haiyan in November 2013. The researchers made
use of primary and secondary data in order to explore humanitarian operations
management challenges before, during and after the onslaught of the typhoon by
analyzing the current humanitarian logistics practices in the country.
3 - Vehicle Management Policies under Stochastic Budget for
Humanitarian Development Programs
Milad Keshvari Fard, ESSEC Business School, 1 Avenue Bernard
Hirsch, Cergy, France, ,
Mahyar Eftekhar, Felix Papier
In this paper we attempt to find the optimal fleet management policies in a
humanitarian organization running development programs. As the budget of
these organizations - mostly financed through donations- is uncertain, the
transportation planning is challenging. We try to identify the optimal number of
vehicles to be purchased, the fraction of demand to be satisfied, and the fraction
of budget that should be saved for future periods with low levels of donations.
4 - A Stochastic Programming Model for Prepositioning and
Distributing Emergency Supplies
Xiaofeng Nie, Assistant Professor, Nanyang Technological
University, 50 Nanyang Avenue, Singapore, Singapore,
Aakil Caunhye, Yidong Zhang,
Mingzhe Li
We propose a two-stage stochastic program to preposition and distribute
emergency supplies. In the first stage the model decides where to locate
warehouses and how many quantities to stock for prepositioning purposes, while
in the second stage the model decides how many quantities to transport to
demand sites and the corresponding routing. A case study is provided to illustrate
our model and compare with other benchmark models.
SB37
37-Room 414, Marriott
Health Care Modeling and Optimization II
Contributed Session
Chair: Dongping Du, University of SOuth Florda, Tampa, FL, 33613,
United States of America,
1 - A Network Heuristic for Stochastic Healthcare Facilities Location
and Configuration in Sequence
Xue Han, Assistant Professor, Southeast Missouri State University,
1933 Rock Creek Ln, Cape Girardeau, MO, 63701,
United States of America, Wilbert Wilhelm
This research aims on finding the best locations and capacities for new healthcare
facilities providing multiple services over a 10-20 years planning horizon under
stochastic demand in a competitive environment. We develop a heuristic
algorithm for introducing new facilities in the market in sequence by solving
resource constrained shortest path problem iteratively on a specially-constructed
network. This heuristic preserves the advantages in solution time and linearity in
objective function.
2 - Simulation Optimization for Reconstructing Rhythmic
Mechanisms in Atrial Fibrillation
Dongping Du, University of SOuth Florda, Tampa, FL, 33613,
United States of America, Hui Yang
Atrial fibrillation (AF) is a common cardiac arrhythmia that affects more than 5
million Americans. The understanding of AF initiation and maintenance has
remained sketchy due to the inability of reconstructing rhythmic mechanisms.
This study develops a multi-scale atrial model and an optimization algorithm to
reconstruct fibrillatory conduction and replicate patterns in clinical recordings.
The research will produce a computer-aided decision support tool for optimizing
AF surgical treatments.
SB37