Operations Management
Message from the Dean
News
FA L L 2 0 1 2
The Operations Management
(OM) field is thriving at the
Smith School of Business. Our
faculty is very well known
in the academic community
and are beginning to be well
known in the practitioner
community as well. The
research ranking of the
Operations Management
faculty is regularly in the
top-10. The faculty are also
associate editors in some of
the most prestigious journals
like Manufacturing & Service
Operations Management
(M&SOM), Production and Operations Management (POM)
and Management Science. Our faculty also work on the most
important issues of the day including healthcare operations,
pricing and revenue management, product innovation, energy/
environmental issues and social networks.
Operations Management is also a well sought after
undergraduate degree, and coupled with supply chain
management is becoming a niche for our MBA students as
well. OM courses now proliferate in our new Master of Science
programs. We believe that with good business principles one can
do a lot of good in this world. Many of the students from the
Smith School of Business work on consulting projects for nonprofit companies, and the single most important toolkit that they
use are concepts and ideas from operations management. We are
very happy that the faculty in the OM area contribute significantly
to the research, teaching and outreach mission of the Smith
School. We are confident that our OM alums will also make the
world a better place.
Dean G. “Anand” Anandalingam
Robert H. Smith School of Business,
Operations Management Faculty
OM Faculty
Frank Alt
Ph.D., Georgia Institute of Technology
Frank Alt is an Associate Professor of Management Science & Statistics. Alt is a recipient of the Krowe Award
for Teaching Excellence (1993 and 1996) and the Distinguished IBM-TQ Teaching Award (1994) from the
Robert H. Smith School of Business. He was recognized with four stars in the Business Week Guide to the Best
Business Schools (1997) for his teaching performance in the MBA program. Alt’s research interests include
statistical quality control, applied multivariate analysis, and forecasting with a particular interest and expertise
in multivariate process control.
Michael Ball
Ph.D., Cornell University
Michael Ball is the Orkand Corporation Professor of Management Science and the Associate Dean of Faculty
and Research. He also holds a joint appointment within the Institute for Systems Research (ISR) in the Clark
School of Engineering. Dr. Ball’s research interests are in network optimization and integer programming
particularly as applied to problems in transportation systems and supply chain management. He is the CoDirector of NEXTOR, the National Center of Excellence for Aviation Operations Research.
Sean Barnes
Ph.D., University of Maryland
Sean Barnes joined the department as an Assistant Professor of Operations Management in Fall 2012. His current
research interests are modeling the transmission of infectious diseases, healthcare operations management,
simulation, and network analysis. He has authored work in peer-reviewed publications such as the INFORMS
Journal on Computing, and Infection Control and Hospital Epidemiology. He has also authored a chapter on
Applications of Agent-Based Modeling and Simulation to Healthcare Operations Management to appear in an
upcoming book.
Margrét Vilborg Bjarnadóttir
Ph.D., Massachusetts Institute of Technology
Margrét Vilborg Bjarnadóttir, is an Assistant Professor of Management Science and Statistics. Prior to joining
Smith, she held an academic position with Stanford Graduate School of Business. She specializes in operations
research methods using large scale data. Her work spans applications ranging from analyzing nation-wide
cross-ownership patterns and systemic risk in finance to drug surveillance and practice patterns in health
care. She has consulted with both health care start-ups on risk modeling using health care data as well as
governmental agencies such as a central bank on data-driven fraud detection algorithms.
Zhi-Long Chen
Ph.D., Princeton University
Zhi-Long Chen is a Professor of Operations Management. Prior to joining the Smith School in 2001, he
was an assistant professor at University of Pennsylvania for 4 years. His research interests cover supply chain
scheduling and coordination, routing/scheduling of logistics operations, capacity planning, and dynamic
pricing. Dr. Chen has conducted a number of NSF funded research projects and is working closely with
industry on several projects in the areas of supply chain optimization and pricing.
Wedad Elmaghraby
Ph.D., University of California at Berkeley
Wedad Elmaghraby is an Associate Professor of Management Science and Operations Management. Prior to
joining the Smith School, she was on the faculty of the ISyE at Georgia Institute of Technology and NYU Stern
School of Business in the Operations Management group. Her current research interests are in behavioral
operations management with an emphasis in (i) pricing and (ii) the design and study of online auctions for
business-to-business markets.
2
Robert H. Smith School of Business
Michael Fu
Ph.D., Harvard University
Michael Fu is Ralph J. Tyser Professor of Management Science. In addition, he has a joint appointment with the
Institute for Systems Research and an affiliate appointment with the Department of Electrical and Computer
Engineering, both in the Clark School of Engineering. He was named a Distinguished Scholar-Teacher at
the University of Maryland for 2004-2005. His research interests include simulation modeling and analysis,
operations management, applied probability and queueing theory, with application to manufacturing and
finance.
Bruce Golden
Ph.D., Massachusetts Institute of Technology
Bruce Golden joined the faculty of the UMD Business School in 1976 and served as a Department Chairman
from 1980 to 1996. Currently, he is the France-Merrick Chair in Management Science at the Robert H. Smith
School of Business at the UMD. He has received numerous awards, including the Thomas L. Saaty Prize (1994
and 2005), the University of Maryland Distinguished Scholar-Teacher Award (2000), the INFORMS Award
for the Teaching of OR/MS Practice (2003), and the INFORMS Computing Society Prize (2005). His research
interests include heuristic search, combinatorial optimization, networks, and applied operations research.
Raghu Raghavan
Ph.D., Massachusetts Institute of Technology
Raghu Raghavan is Professor of Management Science and Operations Management. His research interests
and activities cover a broad domain including- auction design, data mining, economics, information systems,
computational marketing, networks, optimization, and telecommunications. He has published on a wide
variety of topics and numerous academic outlets such as Management Science, Operations Research,
Decision Support Systems, and the INFORMS Journal on Computing. He holds two patents, and has won
numerous awards for his work.
Ilya Ryzhov
Ph.D., Princeton University
Ilya Ryzhov is an Assistant Professor of Operations Management and Management Science. His research deals
with the role of information in decision analysis, exploring the way in which new information influences and
improves decision-making strategies. He develops efficient ways to achieve this balance, with applications
in pricing, revenue management, and optimization of energy costs. His work has appeared in Operations
Research. He is also the coauthor (with W.B. Powell) of the book Optimal Learning.
Tunay Tunca
Ph.D., Stanford University
Tunay Tunca is an Associate Professor of Operations Management and Management Science. Prior to
joining University of Maryland, he was an Associate Professor of Operations, Information, and Technology
at Graduate School of Business at Stanford University. He has also held positions as a visiting scholar at
Wharton, M.I.T., Hewlett Packard and Yahoo Inc. His research interests include economics of operations
and technology management, theoretical and empirical analysis of procurement contracts and processes,
economics of security, and the role of information and forecasting in supply chains.
Yi Xu
Ph.D., University of Pennsylvania
Yi Xu is an Assistant Professor of Operations Management. He teaches and conducts research in the areas of
operations management with an emphasis on product assortment planning, pricing, product development
and innovation, and Marketing and Operations Interfaces. He is particularly interested in exploring
mechanisms that spur innovations and developments of new technologies which transform competitive
dynamics and enable novel operational strategies.
3
Fall 2012 Operations Management News
Selected Recent Publications
M. Ball and M. Queyranne, (2009) “Robust
M. C. Fu, L.J. Hong and J.Q. Hu, (Dec.
M. G. Bardossy and S. Raghavan, (2010)
Revenue Management: A Competitive
2009) “Conditional Monte Carlo Estimation
"Dual-Based Local Search for the Connected
Analysis of Online Booking,” Operations
of Quantile Sensitivities,” Management
Facility Location and Related Problems,"
Research, 57, 950-963.
Science, 55 (12), 2019-2027.
INFORMS Journal on Computing, 22 (4),
Y. Lan, M. Ball and I. Karaesmen,
M. C. Fu, S. Lele and T. Vossen, (2009)
(2011) “Overbooking and Fare-Class
“Conditional Monte Carlo Gradient
Allocation with Limited Information,”
I.O. Ryzhov and W.B. Powell, (2011)
Estimation in Economic Design of Control
Manufacturing & Service Operations
“Information collection on a graph.”
Limits”, Production and Operations
Management, 13, 194-226.
Management, 18 (1), 60-77.
Operations Research, 59 (1), 188-201.
S. Barnes, B. Golden, E. Wasil, (2010)
M. C. Fu, E. Zhou and S.I. Marcus, (2010)
"MRSA Transmission Reduction using Agent-
Firms Invest in Forecasting Efficiently?
“Solving Continuous State POMDPS via
Based Modeling and Simulation," INFORMS
The Effect of Competition on Demand
Density Projection”, IEEE Transactions in
Journal on Computing, 22 (4), 635-646.
Forecast Investments and Supply Chain
Automatic Control, 55 (5), 1101-1116.
Coordination," Operations Research, 58
Z.-L. Chen, (2010) “Integrated production
C. Groer, B. Golden and E. Wasil, (2009)
and outbound distribution scheduling:
"The Consistent Vehicle Routing Problem,"
T. I. Tunca and Q. Wu, (2009) "Multiple
Review and extensions,” Operations
Manufacturing & Service Operations
Sourcing and Procurement Process Selection
Research, 58, 130-148.
Management, 11 (4), 630-643.
with Bidding Events," Management
Z.-L. Chen, and N.G. Hall, (2010) “The
C. Price, M. Harrington, R. Konewko, B.
coordination of pricing and scheduling
Golden, E. Wasil and W. Herring, (2011)
G. Kok and Y. Xu, (2011) "Optimal and
decisions,” Manufacturing & Service
“Reducing Boarding in a Post-Anesthesia
Competitive Assortments with Endogenous
Operations Management, 12, 77-92.
Care Unit,” Production and Operations
Pricing Under Hierarchical Consumer Choice
Management, 20 (3), 431-441.
Models," Management
584-602.
H. Shin and T. I. Tunca, (2010) "Do
W. J. Elmaghraby, S. A. Lippman, C. S. Tang
and R. Yin (2009) “Will More Purchasing
R. Day and S. Raghavan, (2009) "Matrix
Options Benefit Customers?” Production and
Bidding in Combinatorial Auctions,"
Operations Management, 18 (4), 381-401.
Operations Research, 57 (4), 916-933.
(6), 1592-1610.
Science, 55 (5), 763-780.
Science, 57 (9), 1546–1563.
Recent Honors & Awards
Michael Ball
Sean Barnes
• Senior Editor, POM
• INFORMS Fellow
• Finalist for the Pierskalla Award for the
best paper in Healthcare Management
Science (2010)
• Associate Editor, Management Science,
M&SOM
• Received Best Paper Award in Finance
and Policy at the 9th USA/Europe ATM
2011 R&D Seminar
• Co-Director of NEXTOR, National Center
of Excellence for Aviation Operations
Research (2009 - present)
• Member of National Research Council
Panel: Review of Air Traffic Control Enroute
Complexity Model (2009, 2010)
• Area Editor, Operations Research
(2009-2011)
• Associate Editor, Operations Research
4
Robert H. Smith School of Business
Zhi-Long Chen
• Changjiang Scholar Award given by
China’s Ministry of Education (2009)
• Plenary Talk, Tenth International
Conference in Information and
Management Sciences, held in Tibet, China
(2011)
• Associate Editor, Operations Research
Wedad Elmagraby
• NSF Grant: SBE # 0624202: “Collaborative
Research - Consumer Choice and
Organizational Decision-Making” (2008)
Michael C. Fu
• INFORMS Fellow
• IEEE Fellow
• Winter Simulation Conference Best
Theoretical Paper Award (2009)
• OR Program Director, National Science
Foundation (2010-2012)
Bruce Golden
• INFORMS Fellow
• Ranked first among 30 INFORMS
Fellows with respect to h-index and other
measures (2008)
(continued on p.8)
y definition, urgent care patients cannot
B
for capitated insurance systems such as Kaiser
choose when they need to go to the hospital.
Permanete and the Geisinger Health System,
But hospitals have to choose when, and for how
which have a set amount of funding designated
long, to offer certain specialized treatments; to
for each person they cover, regardless of whether
accommodate specialists’ schedules this is usually
that person seeks care or not.
during the day. Since elective surgeries are also
RESEARCH
FOCUS:
WORTH
THE
WAIT
Bruce Golden
usually scheduled during the day, the hospital is
Past studies have tended to focus on one quality
less busy at night. Therefore, a patient visiting
measure, like readmission rates or waiting times,
the emergency room in the middle of night often
and address the issue from a clinical perspective.
gets quicker access to surgery and the intensive
They mostly use annually aggregated national
care unit (ICU). It is well known that long waiting
data, whereas Professor Golden’s study focus is on
times can impair health care outcomes, so it
within-hospital quality variation. The study data
could perhaps be assumed that night patients
includes treatment and outcome measures for
experience comparably improved outcomes.
nearly two million patients and comes from the
American College of Surgeons’ National Trauma
Seeking to identify the mechanisms driving the
Data Bank. The estimation model includes several
variations in care quality for trauma patients,
measures of quality, but the researchers were
France-Merrick Chair Bruce Golden and his
careful to avoid including variables that overlap
colleagues have taken on this question; they find
too much (e.g., the fixed effect of the first trauma
that outcomes are worse at night, so factors other
center and the fixed effect for each center that
than wait times must be having important effects.
is based on the first one), so that the variables’
Professor Golden notes, “There is a ‘Golden Hour’
individual predicting power is preserved. Also,
in medicine. Not named for me, it describes
since the model includes variables that control for
the period – also not necessarily 60 minutes in
the hospital type (level of trauma center, range
length – in which specialized surgery and other
of specialists available), they can show that the
procedures can do the most good for trauma
effect of trauma timing is stronger for lower order
patients. Since night arrivals are not waiting as
trauma centers.
Related Information
David Anderson, Gordon
Gao, and Bruce Golden
“Life is all About Timing: An
Examination of Differences in
Treatment Quality for Trauma
Patients Based on Hospital
Arrival Time”, submitted for
publication.
long for surgery, they are often treated within
David Anderson, Carter Price,
Bruce Golden, Wolfgang
Jank, and Edward Wasil,
(Dec. 2011)“Examining the
Discharge Practices of Surgeons
at a Large Medical Center.”
Health Care Management
Science, 14 (4), 338-47.
hours.”
David Anderson, Bruce Golden,
Wolfgang Jank, and Edward
Wasil, (2012 March)“The
Impact of Hospital Utilization
on Patient Readmission Rate.”
Health Care Management
Science, 15 (1), 29-36.
this period, but their outcomes are still wanting
These trauma centers, which tend to be located
when compared to day patients. This is because
at rural hospitals, need to be even more strategic
senior surgeons tend to work during the day and
with their resources. “Common sense solutions,
the generalists working at night are reluctant to
such as using telecommunications solutions to
call in the eye surgeon at four in the morning,
access specialized guidance and having specialized
for example. Consider also that surgeons may
surgeons work evenings on a regular basis, could
work the night shift only a few times a month
significantly improve outcomes without too much
and therefore are tired, not used to keeping such
extra investment” suggests Professor Golden.
Reforms to hospitals’ internal decision-making
Looking at this problem from the trauma care
process could bring about across the board results
perspective is useful, since the treatment cycle is
by addressing chronic problems (e.g., too few
relatively short and events in the cycle are easy to
specialized surgeons available). This research
represent in a mathematical estimation model (did
offers valuable insights for managers seeking to
the patient require follow-up surgery, 1 for yes, 0
find a mix of hospital resources that can result
for no). It is also important because trauma from
in consistently high quality care. It suggests that
unintentional injury is the number one cause of
daytime results are not better because more care
death among young, wealth-creating individuals.
is available; the degree of care specialization
Improving this situation could have a positive
available also drives variations in care quality
impact on national income, vital at a time of rising
resulting from patient arrival time. This means
health care costs. Furthermore, reducing the extra
that managers cannot just add more care at night;
surgeries that result from temporary “patch”
instead they need to be strategic with the level of
surgeries could result in significant cost savings
resources that they make available.
5
Fall 2012 Operations Management News
C
ompanies are risk-takers: they must make
really of-the-moment data, to update or correct
decisions under uncertain circumstances
his beliefs about an environment,” explains
and, although these have immediate profit
Dr. Ryzhov. “It performs comparably well to
or loss consequences, any result is economic
the technically optimal information-collecting
information that can improve future choices
strategy, while being substantially easier to
(and hopefully increase profits). If a company
implement in practice.” The formula performs
makes a price change and then observes that
well in a variety of situations, but its special value
demand goes down, it can conclude that future
is in application to problems where, for example,
consumers may not be willing to pay as much as
knowing the demand for one product could
they had thought. In industries such as energy
provide demand information for similar products
A New Way
and e-commerce, where prices can change every
(known as “correlated beliefs” in the jargon).
to Think
costs. Firms trading in these markets need a
So with this new tool the online information’s
About
way to learn from the sales fluctuations and
economic value is compared with the revenues
immediately make better decisions.
or costs resulting from the decision; managers
On a smaller scale, a firm that performs energy
can use it to calibrate their desired balance
efficiency upgrades to office buildings has to
of revenue and information. In the near term
choose combinations of upgrades to offer,
it is often worth sacrificing some economic
like both efficient light bulbs and windows,
rewards in order to collect more and better
and online learning can help them find the
online information, which may bring greater
combinations that will record an above-
rewards in the long term. Dr. Ryzhov notes,
average reduction in energy costs and therefore
“The real-time information is especially useful
complement each other particularly well. Firms
in volatile markets, where each decision carries
that want to hedge against high electricity costs
greater risk and uncertainty. This new technique
can buy forward contracts for today’s prices
is important because companies can use it to
and online learning can help them determine
assign economic value to that information and
how many they should buy. In all these cases,
thus compare it to the opportunity costs of
the decisions’ results provide valuable online
making a short-term loss.”
RESEARCH
FOCUS:
Online
Information
Ilya Ryzhov
hour, setting a price too has immediate economic
information in addition to their economic impact.
Online learning requires experimentation, like
Related information
Ilya Ryzhov, Warren Powell
and Peter Frazier, (2012) “The
Knowledge Gradient Algorithm
for a General Class of Online
Learning Problems,” Operations
Assistant Professor Ilya Ryzhov, together with
setting the features in a service and phasing them
Princeton Professor Warren Powell and Cornell
in and out. This experimentation affects revenue,
Assistant Professor Peter Frazier, has developed
especially when it involves lowering prices, but
a new way to assign economic value to online
the collected information calibrates choices
information. Not to be confused with information
that will improve results in the longer term. Dr.
available on the Internet, it is“online” because
Ryzhov’s approach measures to see whether the
the decision-maker observes the information as
value of the new information from the higher
he goes, in addition to the immediate economic
price is enough to compensate for the revenue
benefit (or loss) that occurs as a result of each
penalty. A new weapon in the battle between far
decision. Compare this to the “offline” learning
and near term gains.
Research, 60 (1), 180-195.
that occurs when a product is tested against
Ilya Ryzhov and Warren Powell,
brought to market.
historical data or in the lab setting before being
(2011) “Information Collection
on a Graph," Operations Research,
“Our method mathematically represents how
59 (1), 188-201.
a decision-maker could use online information,
6
Robert H. Smith School of Business
C
RESEARCH
FOCUS:
Innovative
Innovation
Yi Xu
ompanies that have a problem without
a problem is highly complex and a lot of
an obvious solution often turn to their
expertise is needed, contest coordinating
internal research and development (R&D)
costs may be prohibitive. Contest organizers
department, a classic producer of new ideas.
might need to develop a pool of qualified
R&D departments differ from other parts
solvers, for example through advertising or
of a company (e.g., marketing) in that the
sending invitations. In the case of a highly
task of finding one outstanding idea implies
complex problem, it may be cost-effective
many costly failures. A process that has
to hire an intermediary such as Innocentive.
been lowering the cost of poor results is the
com,which advertises the problem to its stable
innovation contest,a form of crowd-sourcing.
of experts. Those who are available to work
In 2010, Google announced the Google Lunar
on the problem submit their solutions to the
X Prize with $30 million in contest prizes to
intermediary and it passes the solutions on to
the first privately funded team to send a robot
the seeking company. Hiring an intermediary
to the moon. In 2006, Netflix launched the
can thus allow the seeker’s identity to remain
Netflix Prize, an open contest with $1 million
private, which can be beneficial in are as like
prizes for the best collaborative filtering
the pharmaceutical industry, where companies
algorithm to predict user ratings for films.
may not want outsiders to know what drugs
Innovative in itself, the innovation contest
they are working on. Limiting the number of
transfers the cost of failure on unsuccessful
solvers in this way also serves to mitigate the
entrants since members of the public submit
under investment of effort problem.
solutions to a posted problem in the hope of
winning a prize.
It has been long thought that the problem of
participants’ under-effort in a large contest
Related information
Christian Terwiesch and Yi Xu,
(2008) “Innovation Contests,
Open Innovation, and
Multi-agent Problem Solving”
Management Science, 54 (9),
1529–1543.
Curtis. R. Taylor, (1995) "Digging
for Golden Carrots: An Analysis of
Research Tournaments" America
Economic Review, 85 (4), 872–890.
But the biggest benefit of a competitive
would mean that running an innovation
innovation contest is not the lowered cost.
contest openly, for example as through a social
Using a model based on extreme-value
network,cannot be a successful R&D strategy.
distribution, Assistant Professor Yi Xu and his
But this research shows that the potential to
coauthor Christian Terwiesch, Andrew M.
obtain revolutionary ideas can easily outweigh
Heller Professor at the Wharton School,have
the under investment of time issue. “When
shown that the wide variety of proposed
we began our research back in 2005,” notes
solutions from outside participants can be
Professor Xu,“Crowd-sourcing and social
predicted to actually produce more diverse
networking were still in their infancy. Our
and creative ideas than an internal R&D
model was able to predict the power of
department. Their work characterizes this
harvesting ideas from an open social network.”
benefit of innovation contest for three
types of innovation. Holding an innovation
contest could actually be more costly than
maintaining a department, both in terms of
coordination costs and also the lowered effort
of participants who think they are unlikely to
win. Yet Xu and Terwiesch find that the benefit
Henry W. Chesbrough, (2003)
"Open Innovation, The New
Imperative for Creating and
Profiting from Technology"
Harvard Business School Press,
Cambridge, MA.
of obtaining more diverse and creative ideas
can easily outweigh these issues.
Carefully designing and managing contests
can further mitigate these issues. Yet when
7
Fall 2012 Operations Management News
4306 Van Munching Hall
University of Maryland
College Park, MD 20742
FA L L 2 0 1 2
Operations Management
News
Recent Honors & Awards
Bruce Golden (continued from p.4)
• Harvey J. Greenberg Award for lifetime
contributions to the INFORMS Computing
Society (2011)
• Finalist for the Pierskalla Award for the
best paper in Healthcare Management
Science (2010)
• Invited to be a Plenary Speaker at the
Annual Meeting of the Italian Association
of Operations Research (2012)
• Finalist for the Best Published Paper in
the last three years in M&SOM (2012)
• Editor in Chief, Networks
• INFORMS Computing Society Prize
for a paper on the use of optimization
techniques to calculate fair payments
in public sector combinatorial auctions.
(2008)
• Winner, Management Science Strategic
Innovation Prize. (2010)
• Area Editor, Informs Journal of Computing
• Doctoral dissertation adviser for
the winner of INFORMS Section on
Location Analysis Biennial Dissertation
Award, and the finalist for INFORMS
Telecomnunications Section Dissertation
Prize (2012).
• Senior Editor, POM
Raghu Raghavan
• INFORMS Telecommunications Section
Outstanding Service Award (2008)
8
Robert H. Smith School of Business
• Doctoral dissertation advisor to the
finalist for George B. Dantzig Dissertation
Prize for the best dissertation in operations
research and management science (2007)
• Glover-Klingman Prize for the best paper
of the year published in Networks. (2006)
Ilya Ryzhov
• Best Paper, IEEE Symposium on Adaptive
Dynamic Programming and Reinforcement
Learning (2011)
Tunay Tunca
• Coauthor for M&SOM Best Student
Paper Finalist Paper (2007)
• Finalist for the Best Published Paper in
the last three years in M&SOM (2009)
Yi Xu
• Management Science Meritorious Service
Award (2010)
• Senior Editor, POM