2016 INFORMS Annual Meeting Program

218 / 561

INFORMS Nashville – 2016
218
4 - Drivers And Benefits Of Return Policies For Online Retailers
Guangzhi Shang, Florida State University,
The common retail practice of full refunds is inconsistent with the
recommendations of many analyticalmodels of returns, which almost always
show that a partial refund is optimal. We use data collected from eBay to analyze
both the return policy drivers from the seller’s perspective and the return policy
value from the consumer’s perspective, providing insights to two potential
explanations for the gap between theory and practice.
MD30
202B-MCC
Retail Operations II
Sponsored: Manufacturing & Service Oper Mgmt
Sponsored Session
Chair: Santiago Gallino, Dartmouth College, Hanover, NH,
United States,
1 - Optimizing Customer Pick-up Locations Using An
Empirical Model
Chloe Kim, The Whaton School, ,
Marshall L Fisher, Xuanming Su
We empirically study the determinants of pick-up location success for an online
grocery retailer. We consider various aspects of individual locations, including
local competition, local consumer attributes, and the potential for cannibalizing
sales from other locations. We suggest an algorithm for optimizing pick-up
locations and measure its impact on revenue.
2 - Prioritizing Outbound Calls In A Sales Contact Center
Marcelo Olivares, Universidad de Chile,
We study an outbound call center that provides sales support to an online
platform selling auto insurance. When setting priorities among which customers
to contact, it is necessary to identify which calls are more likely to convert into
sales. We develop an econometric approach to conduct this segmentation using
detailed customer-level transaction data. A key factor in our analysis is the
customer elapsed time since receiving the quote, which has a significant negative
effect in sales conversion. This approach can be used to implement an automated
priority system to manage outbound calls and can be adapted to other types of
contact centers.
3 - Management And Effects Of In-store Promotional Displays
Oguz Cetin, UNC Kenan-Flagler Business School,
Adam J Mersereau,
Ali Kemal Parlakturk
Promoting a product via an in-store display can increase demand for it by making
it more visible to customers, but it may also impact demand for other products by
changing customer traffic patterns in the store. We characterize the optimal
choice of product to promote in a nested multinomial logit formulation, and we
examine the impact on product, category, and store profits.
4 - Spatial Competition And Preemptive Entry In The Discount
Retail Industry
Fanyin Zheng, Columbia Business School,
This paper studies how discount retailers make store location decisions by
estimating a dynamic game model. It extends the empirical models of dynamic
oligopoly entry by allowing for spatially interdependent entry and introducing
machine learning tools to infer market divisions from data. The results suggest
that preemptive incentives are important in chain stores’ location decisions and
that they lead to loss of production efficiency.
MD31
202C-MCC
Operations and Financing Interface
Sponsored: Manufacturing & Service Oper Mgmt, iFORM
Sponsored Session
Chair: Guoming Lai, Univ. of Texas Austin, 2110 Speedway Stop
B6500, Austin, TX, 78712, United States,
Co-Chair: Qi Wu, Case Western Reserve University, Cleveland, OH,
44106, United States,
1 - Capital Structure With Asset Flexibility
Qi Wu, CWRU, Cleveland, OH, United States,
Peter Ritchken
We study the impact of asset flexibility on the design of an optimal capital
structure in a dynamic model in which the firm has multiple debt issues and
equityholders choose the timing and financing of future growth options as well as
the operating policy for assets in place. We show that, all things being equal,
profitable firms with flexible assets exercise their growth options earlier, use less
debt, and will typically be less leveraged than otherwise identical firms with no
asset flexibility. When asset flexibility allows risk-shifting possibilities, firms
exercise growth options even earlier.
2 - Operational And Financial Interactions In Supply Chain
Network Structure
John R Birge, University of Chicago,
3 - Competitive Risk Management
Danko Turcic, Washington Univ. in St. Louis, ,
Guang Xiao, Panos Kouvelis
This paper provides a new rationale for hedging that is based, in part, on non-
competitive behavior in product markets. We identify a set of conditions, which
imply that a firm may want to hedge. Both operational and financial hedging
strategies are considered.
4 - Inventory Operations Under The Shadow Of Company Stock Price
Guoming Lai, UT-Austin,
Inventory management is one central problem of operations. The classical
inventory theories typically focus on the operational tradeoffs to optimize the
inventory decisions. In practice, however, many firms are public that often have
short-term interests in their market value. Our analysis reveals that public firms
may install more inventory in equilibrium to influence their earnings. We discuss
the determinants of such an effect as well as the empirical evidence.
MD32
203A-MCC
Scheduling
Contributed Session
Chair: Med Labidi, Assistant professor, King Saud University, IED,
Riyadh, 11421, Saudi Arabia,
1 - A Mathematical Model And A Heuristic For The Capacitated Lot
Sizing Problem In Process Industries
Chandrasekharan Rajendran, Professor, IIT Madras, DoMS, IIT
Madras, Chennai, 600036, India, Ramya Ravi,
Hans Ziegler
In this work, we consider the capacitated lot sizing problem (CLSP) with
production carryover and setup carryover across multiple periods in process
industries. This CLSP is characterized by the commencement of production
immediately after the setup of the product, and without any interruption in
production. Mathematical and heuristic models are specifically developed for this
class of CLSP. Also, the proposed heuristic exploits the Fixed-and-Relaxed
principle. The computational effectiveness of both the models are evaluated using
problem instances of various sizes, and are also reported. It is possibly the first
time that such a CLSP is presented.
2 - Scheduling When You Don’t Know The Number Of Machines
Mingxian Zhong, Columbia University, 528 Riverside Drive, Apt
4A, New York, NY, 10027, United States, ,
Clifford Stein
We study scheduling environments in which you have to make some decisions
before learning the number of machines. Specifically, we introduce a model in
which you first need to group jobs into some sets, next you learn the number of
machines and then you need to schedule sets of jobs on machines without
separating them. We give a 9/5(1+epsilon)-approximation algorithm for
minimizing makespan in this environment and show some hardness results.
MD30