2016 INFORMS Annual Meeting Program

TC56

INFORMS Nashville – 2016

3 - Data Analytics Approaches For Winning Service Contracts: Development And Impacts on Practice Aly Megahed, IBM Research - Almaden, San Jose, CA, 95123, United States, aly.megahed@us.ibm.com, Taiga Nakamura, Kugamoorthy Gajananan, Mark Smith, Gregory Heim Service providers in B2B often must prepare bids to win service outsourcing contracts. For high-value IT service outsourcing deals, the solution design and deal pricing process can be a complicated, expensive gamble. We present an approach based in data analytics to automate and customize the often manual practices used to configure such contract proposals. We propose a base model and model enhancements, demonstrating their performance using historical contract data and third-party vendor market data. The actionable methods demonstrate how data analytics tools can enable more effective manager decisions regarding contract proposal solution design and pricing. 4 - Welfare Implications Of Congestion Pricing: Evidence from SFpark Hsin-Tien Tsai, University of California, Berkeley, 1822 Francisco St., Apt 4, Berkeley, CA, 94703, United States, hsintien@berkeley.edu, Pnina Feldman, Jun Li SFpark is a congestion pricing program for street parking implemented in San Francisco. We investigate whether consumers benefit from congestion pricing using data from this program. We build a structural model of consumer search and quantify the change in consumer welfare. 5 - On a Variation Of Two-part Tari Pricing Of Services: A Data Driven Approach Charles Thraves, Massachusetts Institute of Technology, Cambridge, MA, United States, cthraves@mit.edu, Georgia Perakis We present a pricing optimization problem for the data plans of a big satellite firm. First we address the problem of missing data (as reservation prices are not directly observed especially for those who are not current customers). We formulate the price optimization problem as a MIP and develop properties and heuristics in order to solve realistic instances providing analytical lower bounds of their performance. We conclude that with our method the company can increase its profits by more than 10% and outperform the current plans’ prices even under misspecifications of the assumptions. Chair: Hueon Lee, PhD Candidate, University of Arkansas, 4207 Bell Engineering Center, 1 University of Arkansas, Fayetteville, AR, 72701, United States, hueonlee@uark.edu 1 - Long Term Outsourcing Under Stochastic Learning And Information Asymmetry Ting Luo, UT Dallas, 7208 Fair Valley Way, Plano, TX, 75024, United States, tingluo2006@gmail.com We study a firm’s procurement and selling decisions in a multiclass demand and multisupplier inventory system. We show that optimal procurement is driven by multisourcing and intertemporal substitution, while optimal selling is driven by customer segmentation and intertemporal rationing. 2 - Selling Luxury Fashion Online With Social Influences Bin Shen, Donghua University, Xuri Building, 1882 Yanan Road,, Donghua University, Shanghai, China, binshenjerry@gmail.com In the luxury fashion retailing industry, consumers can be categorized into the groups of fashion leader and fashion follower. These two groups influence one another and create social influences in the market. In this paper, we construct an analytical model to examine the effects of demand changes on a luxury fashion supply chain with social influences. We consider the case when the supply chain consists of one supplier and one online retailer providing differentiated services to different groups of consumers. 3 - A Multi-product Dynamic Block Stacking Problem With Deterministic Demand Hueon Lee, PhD Candidate, University of Arkansas, 4207 Bell Engineering Center, 1 University of Arkansas, Fayetteville, AR, 72701, United States, hueonlee@uark.edu, Kelly Sullivan, John A White Block stacking is a commonly used storage method for palletized loads where unit loads are stacked on top of each other and stacks are aligned in storage rows having different depths. In a multi-product storage system each product is assigned to storage rows having a specific depth. As the inventory level changes TC55 Music Row 3- Omni Inventory Management V Contributed Session

for a product, it can be relocated to storage rows having a different depth if relocation minimizes the cost of relocation and the cost of storage space. In our formulation, we require all inventory of a product to be stored in rows having the same depth. With the assumption of a given layout and known inventory cycles, we formulate the problem as a variation of the multicommodity flow problem. TC56 Music Row 4- Omni Managing Sales in On-demand Economy Sponsored: EBusiness Sponsored Session Chair: Michelle Wu, Washington State University, WA, United States, michelle.wu@wsu.edu 1 - We Are On The Way: Analysis Of On- Demand Booking Systems yunny.feng@gmail.com, Guangwen (Crystal) Kong, Zizhuo Wang On-demand platforms such as Uber allow passengers with smartphones to submit trip requests and match them to drivers based on their locations and drivers’ availability. We build a model to analyze the efficiency of such on-demand systems and compare it to systems where people hail taxis on streets. We simulate customers’ waiting time in the two different systems and find that customers’ waiting time with on-demand system can be higher than street hailing. We provide a cap policy that takes advantages of both on-demand system and street hailing in order to minimize customers’ waiting time. 2 - Conform Or To Be Cast Out: Quantifying The Effect Of Platform Endorsement And Consumer Generated Reputation In Online Service Marketplace Demand System Yong Tan, University of Washington, ytan@uw.edu, Jinyang Zheng, Youwei Wang We estimate demands for online service to understand heterogeneous sensitivity to platform endorsement and consumer generated reputation, and to investigate “conform or to be cast out” policy which is applied to force sellers to improve platform endorsement. Our finding shows individuals exhibiting consistent sensitivity to consumer generated reputation, but perceiving platform endorsement differently. With regard to the policy, we find that even though casting out reduce variety of sellers, the negative effect is offset by conforming sellers’ improvement. Furthermore, we find sellers’ further quality escalation in the establishment of new equilibrium, benefiting consumer welfare. 3 - Measuring Consumer Surplus In The On Demand Economy The Case Of Ride Sharing Meng Liu, MIT Sloan, Cambridge, MA, United States, mengliu@mit.edu Uber and Lyft, two pioneer ride-sharing platforms have seen dramatic growth over the last few years. To understand their roles and impacts on the economy, we estimate the consumer welfare of these platforms in a structural demand model for rides. In our model, consumers choose their rides based on price, convenience, brand, and unobserved characteristics. Our identification leverage on the taxi/Uber/Lyft trip records from the New York City, and Uber/Lyft surge price and waiting time at granular location-time levels. We contribute to the understanding of the On-Demand Economy by providing evidence of an increase in consumer welfare due to the fast-growing satisfied instantaneous demand. 4 - Mobility Analytics For Parking Support Michelle Wu, Washington State University, michelle.wu@wsu.edu, Zachary Owen, David Simchi-Levi Ford is developing smart mobility business model as part of the company’s strategic plan to deliver the next level of connectivity, mobility, and customer experience. We develop analytics for dynamic pricing strategies that enable efficient matching of supply with demand. Guiyun Feng, Student, University of Minnesota, 1006 27th Avenue SE, Apt E, Minneapolis, MN, 55414, United States,

321