2016 INFORMS Annual Meeting Program

499 / 561

INFORMS Nashville – 2016
499
WE46
209B-MCC
Revenue Management and Pricing with Consumer
Choice Models
Sponsored: Revenue Management & Pricing
Sponsored Session
Chair: Ruxian Wang, The Johns Hopkins Carey Business School,
100 International Dr, Baltimore, MD, 21202, United States,
1 - Product Line Design And Pricing Under Logit Model
Anran Li, Columbia University,
We study a manufacture who wants to design and price products which are
feature-level combinations. This allows us to design products of different quality
at different prices. We assume that demand for products follows a Logit model
(MNL or NL) that measures utility as the aggregate value of the feature-levels and
the unit production cost is equal to the sum of the component’s wholesale prices.
We show that optimal product configurations give priority to feature levels with
high net value. The K best configurations can be obtained through a greedy
algorithm derived from the K-shortest path problem. This allows us to find an
optimal profit configuration in polynomial time.
2 - Demand Estimation Under The Multinomial Logit Model From
Sales Transaction Data
Tarek Abdallah, New York University, New York, NY, United States,
Gustavo Jose Vulcano
We study an MNL model of demand when customers arrive over time in
accordance to a non-homogeneous Poisson process. We characterize conditions
under which the model is identifiable and our maximum likelihood estimates are
consistent. Then, we propose a maximize-minorize (MM) method for estimating
the model parameters. Through an exhaustive set of numerical comparisons we
conclude that the MM-based estimates are of similar quality to the ones obtained
by state-of-the-art benchmarks, but its computation is orders of magnitude faster.
3 - Pricing And Assortment Management Under New Choice Models
Ruxian Wang, Johns Hopkins Carey Business School, Baltimore,
MD, 21202, United States,
When facing multiple products, a consumer chooses the one with the highest
utility, which depends on the product attributes, its own price, and perhaps the
prices of other products. We characterize the structure of the optimal strategies
and provide efficient exact and approximation algorithms.
WE48
210-MCC
Social Media Collective
Invited: Social Media Analytics
Invited Session
Chair: Lauren Rhue, Wake Forest University, 212 Farrell Hall,
Winston-Salem, NC, 27106, United States,
1 - Communication Of Collective Action Over Time: The Arousal,
Interpretation, And Realization Model
Sung Won Kim, University of Illinois at Urbana Champaign,
, Shaila Miranda
Social media have recently been used by participants in social movements. Twitter
was a key tool in the 2011-2012 protest against the Stop the Online Piracy Act
(SOPA). This research develops a theory of the role of affective and cognitive
mechanisms in online social movements. Tweets from the protest against SOPA
were analyzed using vector autoregression and Granger causality analysis. The
results of the Granger causality were used as empirical data points to build a
theory of communication of collective action over time. Affect was found not
only to spread through the community, but also to influence cognitive
mechanisms. Cognitive mechanisms were used to identify problems and their
solutions.
2 - Who Gets Started On Kickstarter? Demographic Variation In
Crowdfunding Success Rates
Jessica Clark, Stern, , Lauren Rhue
This study examines the variance in success rates across racial groups in the
crowdfunding platform Kickstarter using two novel data sets: the race of subjects
in photos associated with projects (determined using facial recognition software),
and linguistic elements of the project descriptions found to be predictive of the
race of the project photos’ subjects. Even controlling for observable differences in
both project and text characteristics, we find that there are significant racial
differences in success rates. The results have managerial implications for
individual fundraisers and platform designers.
3 - Watch Where You Eat!
Jorge Mejia, University of Maryland,
We focus on how social media analytics can be incorporated into the efforts to
reduce foodborne illness, a major public health concern. We demonstrate how
machine learning techniques can be used to monitor the hygiene quality of
restaurants through social media. Using these techniques, we show evidence of
moral hazard in how restaurants achieve high grades in the New York City
program.
WE49
211-MCC
Teaching OR at Service Academies
Sponsored: Education (INFORMED)
Sponsored Session
Chair: Richard McGrath, United States Naval Academy, 121 Blake
Road, Annapolis, MD, 21402, United States,
1 - Teaching Operations Research At The United States
Naval Academy
Richard G McGrath, United States Naval Academy,
The United States Naval Academy (USNA), founded in 1845, is the undergraduate
college of the U.S. naval service. Midshipmen attend the academy for four years,
graduating with Bachelor of Science degrees and commissions as ensigns in the
U.S. Navy or second lieutenants in the U.S. Marine Corps. Since 2012, the
Mathematics Department has offered an undergraduate major in Operations
Research, which has become one of the most popular majors at the academy. In
this talk, we will discuss the OR program at USNA. We will describe a typical
course of study, as well as how military and security applications are incorporated
into the curriculum through coursework, project-based learning, and capstone
projects.
2 - Teaching Operations Research At The United States
Air Force Academy
Gerry Gonzalez, United States Air Force Academy, Colorado
Springs, CO, United States, , Jesse Pietz
The United States Air Force Academy (USAFA) is an undergraduate college and
one of three commissioning sources for the United States Air Force. Cadets attend
the Academy for four years, graduating with Bachelor of Science degrees and
commissions as second lieutenants in the United States Air Force. USAFA has
offered an undergraduate major in Operations Research since 1966. In this talk,
we will discuss the OR program at USAFA. We will describe a typical course of
study, as well as how operational military applications are incorporated into the
curriculum through coursework, project-based learning, and capstone projects.
3 - Teaching Operations Research At The United States
Coast Guard Academy
Melinda D. McGurer, Head, Department of Mathematics, United
States Coast Guard Academy, 15 Mohegan Avenue, New London,
CT, 06320, United States, ,
Eric C. Johnson
The United States Coast Guard Academy (USCGA), founded in 1876 as the
Revenue Cutter School of Instruction, is the undergraduate college of the U.S.
Coast Guard in the Department of Homeland Security. Cadets attend the academy
for four years, graduating with Bachelor of Science degrees and commissions as
ensigns in the U.S. Coast Guard. Since 1997, the Mathematics Department has
offered a degree in operations research. This talk discusses the OR program at the
USCGA including a typical course of study and the incorporation of USCG
applications. The talk emphasizes the senior capstone course in which the cadets
serve as OR consultants for the USCG and broader military and homeland
security communities.
WE49