2016 INFORMS Annual Meeting Program

140 / 561

INFORMS Nashville – 2016
140
2 - A Modified Age Replacement Model
Maryam Hamidi, University of Arizona,
The classical age replacement model is revisited when the scheduled arrival time
of the replacement is also a decision variable in addition to the scheduled
preventive replacement time. The equipment is replaced as scheduled unless it
breaks down earlier, in which case replacement is performed immediately. If the
replacement arrives before it is needed, then inventory cost arises, and if it is not
available when needed then shortage cost of delayed production is the loss of the
company. The production cycle is repeated without any change, therefore the
expected net cost per unit time is computed by renewal theory. The existence of
finite optimum is proved and the computer method is introduced to find it.
3 - Maximizing Availability In Operations And Maintenance Planning
Of Airline Fleet
Javad Seif, Graduate Research Assistant, University of Tennessee-
Knoxville, Tullahoma, TN, 37388, United States, ,
Andrew Junfang Yu
In this paper, operations and maintenance planning (OMP) is modeled as a
mixed-integer linear program with a case study in the airline industry. The
objective is to maximize the availability of a fleet of aircraft over a planning
horizon subject to their operations and maintenance requirements. An existing
flight and maintenance planning (FMP) model is significantly extended by
generalizing it to an OMP model, which makes the model more practical and
suitable for a wider range of applications in both manufacturing and service
industries.
MA56
Music Row 4- Omni
Economics of IS in the Age of Big Data
Sponsored: EBusiness
Sponsored Session
Chair: Wael Jabr, Georgia State University, 35 Broad Street, Robinson
College of Business, Atlanta, GA, 30303, United States,
1 - Antecedents And Consequences Of Information Augmentation In
Make-to-order Electronic Supply Chain
Ling Xue, Georgia State University,
Arun Rai, Peijian Song, Cheng Zhang
In this study, we consider a phenomenon of information augmentation: in
addition to passing the demand information from the consumers to the supplier,
the retailer generates ad hoc coordination information to direct the order-
fulfilment processes at the supplier side. Using transaction-level data from an
international make-to-order supplier chain, we find that information
augmentation is positively associated with product customization level and
negatively associated with supplier relationship. We also find that information
augmentation leads to higher fulfilment costs and lower probabilities of product
return, and higher net profits for the retailer.
2 - Opening The Black Box: The Impact Of Telehealth And Latent
Health Status On Patient Readmissions
Sezgin Ayabakan, The University of Baltimore, Baltimore, MD,
21201, United States,
Indranil R Bardhan, Zhiqiang Eric Zheng
We investigate the effect of unobserved patient health status on patient
readmission rates and the impact of telehealth on patient health status. We
implement a hidden Markov model with a large, inpatient panel dataset of
Congestive Heart Failure patient visits along with the American Hospital
Association IT Supplement data. We reveal latent health states of patients and
find that telehealth exerts a positive impact only on patients in less healthy states,
while this impact diminishes as patients’ health improves. These results suggest
that focusing solely on readmission rates can be misleading, without regard to
patients’ health status.
3 - Leveraging Customer Feedback Through App Reviews
Kambiz Saffarizadeh, Georgia State University, Atlanta, GA, United
States, , Wael Jabr, Mark Keil
Online app markets (e.g. Apple App Store) exhibit heavy customer engagement
in the form of reviews that could help software developers adjust to user needs
and become more competitive in crowded app markets. Using a panel dataset on
12,231 apps and document similarity methods, we develop a model that relates
app success to developers’ integration of user feedback.
4 - Crisis Management Of Customer Sentiments In Social Media:
Empirical Study
Kyungsun Rhee, University of Washington, 4725 24th Avenue NE,
# 405, Seattle, WA, 98105, United States,
In this paper, we will examine how firms strategically bundle news reports in
order to intervene diffusion patterns of negative sentiments in social media after
the event occur in companies. By conducting text mining and estimating unique
data set on Twitter empirically, this paper will contribute to literature by providing
managerial insights of ways to control social media.
MA57
Music Row 5- Omni
The George B. Dantzig Dissertation Prize
Invited: The George B. Dantzig Dissertation Prize
Invited Session
Chair: Henri Groenevelt, University of Rochester, W. E. Simon
Graduate School of Business Administration, Rochester, NY, 14627,
United States,
1 - Dynamic Monitoring And Control Of Irreversible Chronic Diseases
With Application To Glaucoma
Pooyan Kazemian, Harvard Medical School, Boston, MA,
United States,
To effectively manage chronic disease patients, clinicians must know (1) how to
monitor each patient (i.e., when to schedule the next office visit and which subset
of tests to take), and (2) how to control the disease (i.e., what levels of
controllable risk factors will sufficiently slow disease progression). Our research
addresses these questions simultaneously and provides the optimal solution to a
novel linear quadratic Gaussian state space model. Glaucoma is discussed as a case
study.
2 - Multiple Criteria Decision Engineering To Support Management In
Military Healthcare And Logistics Operations
Nathaniel D. Bastian, Pennsylvania State University,
137 Arbor Bluff Drive, Pleasant Gap, PA, 16823, United States,
The U.S. Military Health System (MHS) is a unique, complex health system with
many healthcare and logistics challenges requiring effective management. We
introduce multiple criteria decision engineering methods to assist strategic
decision-making and to support the complex planning and management of MHS
resources, personnel, logistics, and financial incentives. This contribution has
helped senior military leaders make better, more informed, quicker, data-driven
resource management decisions amongst conflicting objectives, along with the
risks and uncertainties associated with those decisions.
3 - Smarter Tools For (Citi) Bike Sharing
Eoin O’Mahony, Uber, Itaca, NY, United States,
Bike-sharing systems are becoming increasingly prevalent in urban environments.
These systems generally consist of stations where users can take out or drop off
bikes, and may return a bike to a free dock at any station. New York City
launched the largest bike sharing system in North America, Citibike, in May 2013
with over 300 stations and 5000 bikes. We have worked with Citibike since
launch, using analytics and optimization to change how they manage the system.
In this talk we will cover two areas of our work with Citibike; planning through
data and rebalancing with optimization.
4 - Optimizing A Menu Of Multi-format Subscription Plans For
Ad-Supported Media Platforms
Vamsi Krishna Kanuri, University of Miami, 501 Kosar Epstein
Bldg, School of Business Administration, Coral Gables, FL, 33146,
United States,
Media content distribution has changed extensively in the past decade. Content,
which was once distributed through traditional formats such as television, radio
and print, is now available through contemporary digital formats such as
smartphone and tablet apps, with many possible versions (e.g., presence or
absence of ads.) Consequently, many media firms facing markets comprised of
heterogeneous consumers with varying content consumption preferences are
now offering ‘menus’ of multi format-version subscription bundles for their
consumers to choose from. Yet, little systematic model-based guidance exists for
configuring and pricing menu options. Moreover, most media firms are ‘audience-
building platforms’ that serve at least two distinct customer groups (content
consumers and advertisers) with inter-related demands. Therefore, constructing a
menu of content subscription bundles that maximizes total profit from both
consumers and advertisers is a formidable challenge. This research proposes a
theory-driven implementable model-based approach that can aid media platforms
in addressing this challenge.
MA56