Informs 2015

258 / 552

INFORMS Philadelphia – 2015
256
4 - Joint Burn-in and Imperfect Condition-based Maintenance
for N-subpopulations
Yisha Xiang, Assistant Professor, Lamar University,
2626 Cherry Engineering Building, Beaumont,, TX, 77710, United
States of America, David Coit
For some engineering design and manufacturing applications, particularly for
evolving and new technologies, some populations of manufactured parts or
devices are heterogeneous and consist of a small number of different
subpopulations. In this study, we propose a joint burn-in and imperfect condition-
based maintenance model with consideration of random effects within
subpopulations. Numerical examples are provided to illustrate the proposed
procedure.
MD75
75-Room 204B, CC
Deep Dive on Open Innovation –
Papers and Discussants
Cluster: New Product Development
Invited Session
Chair: Jeremy Hutchison-Krupat, Professor, University of Virginia,
Charlottesville, VA 22901, United States of America,
1 - Optimal Shapes of Innovation Pipelines
Joel Wooten, University of South Carolina, Columbia, SC,
United States of America, ,
Sriram Venkataraman
New product introductions often occur via R&D pipelines. We explore the optimal
number of innovation options to pursue in this complex managerial process. A
stylized game simulation of the pharma industry provides additional evidence for
our problem.
2 - Discussant
Sanjiv Erat, UCSD, Gilman Drive, La Jolla, CA,
United States of America,
This talk will offer a discussion/critique of the paper titled “Optimal Shapes Of
Innovation Pipelines.
3 - How Much Better is Open Innovation?
Sebastian Fixson, Babson College, Tomasso Hall 226, Babson Park,
MA, 02457, United States of America, ,
Tucker Marion
Over the past 15 years research has emerged that describes many advantages of
open innovation, such as unearthing ideas that better match customer needs
and/or problem specifications. In this paper, we study in detail the new product
development process of a single organization that makes extensive use of external
actors throughout its process, and explore the corresponding performance
implications.
4 - Discussant
Yi Xu, Associate Professor, Smith School of Business, University of
Maryland, College Park, MD, 20742, United States of America,
This talk will offer a discussion/critique of the paper titled “How Much Better Is
Open Innovation?
MD76
76-Room 204C, CC
Simulation in Healthcare
Sponsor: Simulation
Sponsored Session
Chair: Tahir Ekin, Assistant Professor, Texas State University,
01 University Dr. McCoy Hall 411, San Marcos, TX, 78666,
United States of America,
1 - Simulation of Hospital Outpatient Clinics
Lawrence Fulton, Assistant Professor, Texas Tech University,
703 Flint Ave, Lubbock, TX, United States of America,
, Nathaniel Bastian
MedModel was used to provide decision support for a hospital’s outpatient clinic
organization. Variables of interest included cost, capitation rate, utilization, and
throughput. Outpatient areas evaluated included primary care clinics, OB/GYN,
pediatrics, internal medicine, same-day surgery, orthopedics, psychology /
psychiatry, social work service, and physical therapy / occupational therapy. The
modeling demonstrates the usefulness of healthcare simulation for organizational
change.
2 - The Use of Lindley’s Entropy in Dynamic Sampling Decisions
Rasim Muzaffer Musal, Associate Professor, Texas State
University, 601 University Dr., McCoy Hall 411, San Marcos, TX,
78666, United States of America, , Tahir Ekin
Neyman Allocation (NA) is used to stratify Medicare payments to create relatively
homogeneous strata. These strata are assumed to provide a relatively more
homogeneous over-payment sub-populations. We suggest an extension to NA by
the use of Lindley’s expected information gain measure to make efficient
sampling decisions. In doing so a novel application is presented under simulated
scenarios. A comparison between alternative methods is illustrated.
3 - Using Markov Chain Monte Carlo for Input Models of Surgery
Duration in a Multi-specialty Department
Louis Luangkesorn, Research Assistant Professor, University of
Pittsburgh, 1048 Benedum Hall, Department of Industrial
Engineering, Pittsburgh, PA, 15261, United States of America,
Zeynep Filiz Eren Dogu
The variety of procedures in a surgery suite means that even with several years of
data many surgical cases will have little or no historical data for use in predicting
case duration. Parameterizing duration is needed for other procedures such as
stochastic optimization. We combine expert judgement, expert classification of
procedures by complexity category and historical data in a Markov Chain Monte
Carlo (MCMC) model to parameterize cases and test the result against other
methods.
4 - Medicare Fraud Analytics using Cluster Analysis
Babak Zafari, The George Washington University School of
Business, 2201 G Street NW, Funger Hall, Suite 415, Washington,
DC, United States of America, Paulo Macedo,
Sewit Araia
In this work, we use of cluster analysis to group healthcare providers based on
similar billing patterns. In detecting outliers, comparing providers based on self-
reported specialty can cause false positives due to specialization. We use BETOS
codes to categorize procedure codes in addressing the issue of aberrant billing
behavior. This establishes a representative peer comparison group that minimizes
false positives. The efficacy of the proposed method is illustrated through data
simulation.
MD77
77-Room 300, CC
Supply Chain Management VIII
Contributed Session
Chair: Marcus Bellamy, Assistant Professor, Boston University
Questrom School of Business, 595 Commonwealth Avenue, Boston,
MA, 02215, United States of America,
1 - R and D Modes of Manufacturers’ Cost Reduction:
How to Invest in Supply Chains
Jing Hu, PhD Student, Fudan University, 670 Guoshun Road,
Yangpu District, Shanghai, China,
Qiying Hu
Inspired by the Chinese mobile phone industry, we find four R&D modes between
vertical firms: two collaborative modes (R&D cartel and R&D joint venture) and
two non-collaborative modes (manufacturer-R&D and retailer-R&D). A three-
stage game model is considered to explain why these modes coexist. We find that
firms prefer the R&D cartel if and only if they have comparable channel powers.
When collaboration is impossible, only the firm with sufficiently smaller cost
factor prefers R&D by itself.
2 - The Role of Customer Flexibility in Achieving Supply Chain Agility
Vahid Ghomi, PhD Student, University of Mississippi, Marketing
Department, School of Business Administration, Oxford, MS,
38655, United States of America, ,
Bahram Alidaee
A firm’s supply chain agility (SCA) is a critical factor affecting its overall
competitiveness. To create SCA, most research concentrate on exploring
manufacturing flexibility, supply side flexibility, and logistics capabilities.
However, there are variety of settings where demand side flexibility (DSF) can be
achieved. The purpose of this research is to present, (1) a comprehensive
literature review of DSF, (2) research directions as how SCA can be achieved by
exploring DSF.
MD75