2015 Informs Annual Meeting

SC50

INFORMS Philadelphia – 2015

SC49 49-Room 105B, CC Operations Management and Marketing Sponsor: Manufacturing & Service Oper Mgmt/Supply Chain Sponsored Session Chair: Xingwei Lu, University of Pennsylvania, The Wharton School, Philadelphia, United States of America, xingwei@wharton.upenn.edu Co-Chair: Xuanming Su, The Wharton School, University of Pennsylvania, Philadelphia, United States of America, xuanming@wharton.upenn.edu 1 - A Tractable Discrete Choice Model - Beyond Logit Aydin Alptekinoglu, Penn State, Smeal College of Business, University Park, PA, 16802, United States of America, aydin@psu.edu, John Semple We explore the mixed (random coefficients) and heteroscedastic versions of the Exponomial Choice (EC) model. EC has attractive analytical features like closed- form choice probabilities, logconcave likelihood, and more realistic cross-price elasticities compared to multinomial logit. Allowing heteroscedasticity in error terms, and allowing random coefficients with normal distribution, retain closed- form choice probabilities but not some of the other properties. 2 - How Should a Retailer Share Demand Information with the Supplier? Qi Annabelle Feng, Professor, Purdue University, 100 S. Grant St., West Lafayette, IN, United States of America, annabellefeng@purdue.edu, J. George Shanthikumar We consider a supplier selling through a retailer who possesses private demand information. The retail can choose to share a Bayesian plausible demand signal with the supplier. We define the notion of information accuracy and analyze the structure of demand information to understand the retailer’s incentive of information sharing and its effect on supply chain performance. 3 - Minimum Advertised Price Policy: Economic Analysis and Implications Ozge Sahin, Johns Hopkins University, ozge.sahin@jhu.edu, Liang Ding, Roman Kapuscinski Manufacturers frequently use Minimum Advertised Price (MAP) to protect retailers’ margin and to encourage them to exert more sales effort. This paper analyzes the performance of MAP under various market situations and compares it with another vertical price restraint policy – Resale Price Maintenance. Results indicate that MAP could be beneficial to manufacturers and also to customers when the information search is costly and the consumers are moderately heterogeneous in their valuation. 4 - Revenue Management with Loyalty Programs Xingwei Lu, University of Pennsylvania, The Wharton School, Philadelphia, United States of America, xingwei@wharton.upenn.edu, Xuanming Su We study loyalty programs in firms with limited capacity. Based on the classic Littlewood’s model, our model additionally reserves some capacity for awards and allows customers to choose between paying with cash and redeeming with points. We investigate the optimal capacity allocations and compare program designs. SC50 50-Room 106A, CC Procurement and Management of Complex Goods and Services Sponsor: Manufacturing & Service Operations Management Sponsored Session Chair: Damian Beil, Associate Professor of Technology & Operations, Ross School of Business, University of Michigan, 701 Tappan St, Ann Arbor, MI, 48109, United States of America, dbeil@umich.edu 1 - Multi-channel Service with Heterogeneous Customers Shiliang Cui, Georgetown University, McDonough School of Business, Washington, DC, 20057, United States of America, shiliang.cui@georgetown.edu, Yong-Pin Zhou When firms offer multiple service channels, it gives customers a chance to choose the channel that better fits their need (cost, convenience, advice, etc.). On the other hand, firms can also use carefully designed multiple service channels to achieve customer segmentation. We study mechanisms to achieve this.

3 - Feed-in Tariff Versus Rebate for Renewable Generation Ruben Lobel, Operations and Information Management Department at The Wharton School of the University of Pennsylvania, Jon M. Huntsman Hall 3730 Walnut Street, Office 568, Philadelphia, PA, 19104, United States of America, rlobel@wharton.upenn.edu, Vlad Babich We compare two types of subsidies used to stimulate renewable energy generation: a feed-in tariff and an upfront rebate. The former guarantees a steady cash flow over the long term, while the latter reduces the upfront capital requirement. We show how uncertainty and strategic customer behavior determines policy efficiency. 4 - Energy Efficiency and Demand Response on a Production Line Eric Webb, Graduate Student, Indiana University, 1309 E. 10th Street, Bloomington, IN, 47405, United States of America, ermwebb@indiana.edu, Owen Wu Demand response (DR) programs incentivize industrial firms to halt production during times of peak electricity demand. We consider a firm faced with the option of investing in energy efficiency (EE) improvements at individual machines on the production line. When viewed in isolation, EE incentives may not be enough to induce the firm to invest in the socially optimal level of EE, due to the loss of DR revenue after installation. We suggest a new policy for EE incentives in light of DR. Energy and Commodity Merchant Operations Sponsor: Manufacturing & Service Oper Mgmt/iFORM Sponsored Session Chair: Nicola Secomandi, Associate Professor, Carnegie Mellon Tepper School of Business, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States of America, ns7@andrew.cmu.edu Co-Chair: Selvaprabu Nadarajah, Assistant Professor Of Information And Decision Sciences, College of Business, University of Illinois at Chicago, 601 S Morgan Street, 24th Floor, Chicago, IL, 60607, United States of America, selvan@uic.edu 1 - Electricity Derivative Trading with Private Information on Price Distributions Eddie Anderson, University of Sydney, Business School, Sydney, Australia, edward.anderson@sydney.edu.au, Andy Philpott When trading forward contracts, firms try to hedge risk, but at the same time want to benefit from superior price forecasting. We model a negotiation between generator and retailer on forward contracts when both have private information on the probability of different price outcomes. Both firms submit a schedule of prices and quantities, and the market clears. Can a firm gain by looking at the offers that the other player makes in order to improve its own estimate of the real probabilities? 2 - Optimal Production and Shortfall Hedging Liao Wang, Department of Industrial Engineering and Operations Research, Columbia University, S.W. Mudd Building, Room 315, 500 West 120th Street, New York, NY, 10027, United States of America, lw2489@columbia.edu, David D. Yao Over a given time horizon, we study the one-time production quantity decision at the beginning, and the real-time risk-hedging strategy throughout the horizon. The objective is to minimize the shortfall from a given target profit, and we derive the joint optimal solution to both production and hedging. In addition, we characterize the efficient frontier, and quantify the improvement in risk-return tradeoff achieved by the shortfall hedging. 3 - Least Squares Monte Carlo: Duality Perspective and Energy Real Option Application Selvaprabu Nadarajah, Assistant Professor Of Information And Decision Sciences, College of Business, University of Illinois at Chicago, 601 S Morgan Street, 24th Floor, Chicago, IL, 60607, United States of America, selvan@uic.edu, Nicola Secomandi Least squares Monte Carlo (LSM) is a common approximate dynamic programming approach. We provide a duality perspective on a version of this approach, which leads to an equivalent formulation of this approach as a relaxed approximate linear programming (ALP) model. We provide both theoretical and numerical support for the use of LSM rather than ALP. Our numerical study deals with merchant ethanol production, an important energy application. SC48 48-Room 105A, CC

109