2015 Informs Annual Meeting

WB29

INFORMS Philadelphia – 2015

WB28 28-Room 405, Marriott Economics of Auctions Cluster: Auctions Invited Session

4 - Lot Sizing on Multiple Lines with a Single Setup Operator Dirk Briskorn, University of Wuppertal, Rainer-Gruenter-Str. 21, Lehrstuhl för Produktion und Logistik, Wuppertal, 42119, Germany, briskorn@uni-wuppertal.de We consider a lot sizing problem with multiple production lines. There is a single setup operator who needs to be present at a line during setup operations. The lines are otherwise independent and underlie the characteristics of the capacitated lot sizing problem. We present an heuristic approach focussing on the setup sequence of the setup operator. We outline a method yielding initial solutions to be fed to a neighbourhood search algorithm. Finally, we provide first computational results. 5 - Stochastic Lot-sizing and Cyclical Scheduling with Sequence Dependent Setup Time: The Case of Wine We modified a dynamic vehicle routing formulation to determine optimal cyclical schedules and lot sizes with a stochastic component. Which is: variability of setup times and the probability of a certain product or SKU to require production on a given planning horizon is added into the formulation. Solutions will be benchmarked using the variability (variance) of the optimized solution and two different metrics for the objective function: first, line utilization and second, minimize total costs. WB27 27-Room 404, Marriott Applications and Methodological issues on MCDM Sponsor: Multiple Criteria Decision Making Sponsored Session Chair: Danielle Morais, Assistant Professor, Universidade Federal de Pernambuco, Management Engineering Department, Recife, PE, Brazil, daniellemorais@yahoo.com.br Co-Chair: Adiel T De Almeida, Professor, Universidade Federal de Pernambuco, Caixa Postal 7462, Recife, PE, 50630-971, Brazil, almeidaatd@gmail.com 1 - Subcontractor Selection in Civil Construction Most of the activities in construction projects are carried out by subcontractors. Based on this, it is essential for the contractor to have a structured methodology for selecting them. In this study, it is presented a model for subcontractor selection using an additive-veto multicriteria model, in which the contractors can incorporate the veto notion for those subcontractors’ criteria performances that he thinks that is unacceptable. An application is presented. 2 - An Mcdm Model to Assist Preventive Maintenance for a Water Well Equipment in a Low Income Community Adiel De Almeida Filho, Assistant Professor, Universidade Federal de Pernambuco, Caixa Postal 7471, Recife, PE, 50630-971, Brazil, adieltaf@googlemail.com, Madson Monte This work presents a multiple criteria decision model for the preventive maintenance planning of a water well equipment located at a low-income community in Brazil. This specific problem has been addressed considering the local context and limitations in order to find a compromise between cost and service level when defining the time interval for preventive maintenance. 3 - PROMETHEE Method with Surrogate Weights for Group Decision Making Danielle Morais, Assistant Professor, Universidade Federal de Pernambuco, Management Engineering Department, Recife, PE, Brazil, daniellemorais@yahoo.com.br, Luciana Alencar, Tharcylla Negreiros, Adiel T De Almeida A large number of multicriteria group decision models have been developed, differing from each other the quality and quantity of information required. PROMETHEE has been applied in many group decisions, mainly when it is possible to define criteria weights. However, establishing weights can be hard and in some cases is necessary deal with imprecise information regarding the importance of criteria. Thus, we propose a group decision model, integrating the PROMETHEE method with surrogate weights. Luciana Alencar, UFPE, Rua Acadeímico Hélio Ramos, s.n., Recife, Brazil, alencarlh@gmail.com, Adiel T De Almeida Alejandro Mac Cawley, Assistant Professor, P. Universidad Catolica de Chile, Vicuna Mackenna 4860 Macul, Santiago, 7820436, Chile, amac@ing.puc.cl

Chair: Lin Hao, University of Notre Dame, 351 Mendoza College of Business, Notre Dame, IN, United States of America, lhao@nd.edu 1 - Who Benefits from Free Lunch? An Examination of Return Policies at Private Fashion Sale Sites Yixin Lu, Assistant Professor, VU University Amsterdam, Netherlands, y2.lu@vu.nl, Alok Gupta, Wolf Ketter This paper examines the role of return policy at private fashion sale sites (PFSS) that offer limited-time deals of fashion merchandise to registered customers. Compared to traditional online fashion retailers, PFSS often employ more strict return policies to minimize the product return rate. We propose a simulation framework to compare the effects of different return policies on seller’s revenue. 2 - The Cost of Running an Auction and Implications to the Optimal Reserve Price Zohar Strinka, PhD Candidate, University of Michigan, 1205 Beal Ave., Ann Arbor, MI, 48109, United States of America, zstrinka@umich.edu A common question in the auction literature is whether a seller can credibly commit to keep a good if the bidders fail to meet the reserve price. This paper shows that by modeling the seller’s problem as a multi-round auction with reserve prices in each round, we can distinguish an inability to commit from failure to model the full problem. In addition to our parametric analysis, this modeling framework opens the door to a more sophisticated understanding of the costs of a failed auction. 3 - Understanding the Effect of Jump Bidding in Overlapping Online Auctions Lin Hao, University of Notre Dame, 351 Mendoza College of Business, Notre Dame, IN, United States of America, lhao@nd.edu, Yong Tan, Arvind Tripathi In electronic markets, overlapping online auction has emerged as a viable and efficient mechanism for large retailers desiring to clear off their excessive inventories. In this paper, using a dataset from a reputable website we empirically studied the effect of jump bidding behavior on the outcomes of overlapping online auctions. WB29 29-Room 406, Marriott Data Analytics in Pricing and Revenue Management Chair: Jian Wang, Vice President, Research & Development, The Rainmaker Group, 4550 North Point Parkway, Alpharetta, GA, 30022, United States of America, jwang@letitrain.com 1 - Competitive Rate Information: Use and Misuse in Hospitality Revenue Management Alex Dietz, Advisory Industry Consultant, SAS Institute, 100 SAS Campus Drive, Cary, NC, 27513, United States of America, Alex.Dietz@sas.com In today’s Big Data environment, most hotels that are actively managing their pricing use a competitive rate shop mechanism as a vital tool to assist them in pricing. But simply following the lead of competitors can lead to a significant loss in revenue. This presentation will discuss the importance of rate information, how this information can be used in revenue management, and common traps associated with use of this data – including data quality, limitations, and misuses of the information. 2 - Making Television Ad Inventory More Valuable through Optimization Bob Bress, Sr. Director, Prod. Mgmt & Analytics, Visible World, 460 W. 34th Street, NY, NY, 10001, United States of America, bob.bress@visibleworld.com Approaching TV advertising programmatically with advanced data and dynamic decisioning provides the ability to add tremendous value to TV inventory. In this session we will cover: ïHow optimization methods are being deployed to enhance the value of TV inventory through dynamic decisioning ïHow analytics are allowing advertisers to benefit from advanced data through more targeted advertising ïHow both inventory holders and advertisers are benefiting through new data-driven applications for TV. Sponsor: Analytics Sponsored Session

409