Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

WC13

5 - Individualized No-show Predictions: Effect on Clinic Overbooking and Appointment Reminders Yutian Li, University of Science and Technology of China, Sammi Tang, Joseph Johnson, David Lubarsky In this paper, we develop a Bayesian nested logit model which improves no-show prediction accuracy over the widely-used standard logit model. The accuracy gain arises from the individual patient-level coefficients provided by the Bayesian approach. Comparison of model fit on 12-months of appointment data shows the Bayesian model outperforms the standard logit model. In simulation studies, our results show that applying Bayesian model’s prediction to scheduling algorithm can increase clinic’s profit, reduce waiting cost when no-show probability is low and reduce overtime cost when no-show probability is high. n WC11 North Bldg 125B Service Operations Management in the IT Industry Sponsored: Manufacturing & Service Oper Mgmt Sponsored Session Chair: Aly Megahed, IBM, San Jose, CA, United States Kaan Unnu, Rensselaer Polytechnic Institute, Center for Industrial Innovation, 110 8th Street, Troy, NY, 12180-3590, United States, Jennifer A. Pazour, Aly Megahed, Pralhad Deshpande, Chandra Narayanaswami Blockchain technology has the potential to increase supply chain transparency and traceability by creating a robust and trusted decentralized source of information between different supply chain entities. We create a framework to categorize blockchain supply chain use cases based on a supply chain’s echelons and scale, as well as information requirements. The presented framework systematically links potential blockchain benefits and use cases to existing supply chain literature and identifies new directions for future ORMS research topics. 2 - Optimal Bidding for Highly-Valued IT Service Contracts: Theoretical Results and Practical Implications Xiangyu Zhang, Cornell University, Ithaca, NY, United States, Aly Megahed, Peter Frazier Information technology service providers compete to win highly-valued IT service contracts in a tender process. Prior literature shows that features other than price, including the service provider’s relationship with the client, contribute to the client’s selection because good relationship increases the provider’s chance of winning the deal. Thus, it might be beneficial for the provider to lower their price for improving the client relationship and increasing their potential future contracts profits. In this work, we provide theoretical and numerical results illustrating the optimal price is lower than the myopic price which tries to maximize the expected profit of the current deal. 3 - An Approach to “Predict+optimize” Rather than “Predict Then Optimize” with a Case Study in a Cloud Computing Application Aly Megahed, IBM Research-Almaden, San Jose, CA, 95123, United States, Yingdong Lu, Mark S. Squillante In a lot of applications, some uncertain parameters are predicted then fed into an optimization model to optimize for some task. For example, customer demand in a supply chain is predicted and then fed into an optimization model to find the optimal inventory levels. The traditional approach is to do the two tasks separately. We present a novel approach of doing them simultaneously (predict+optimize rather than predict then optimize) and apply it to a real-world service operations management application in cloud computing. 4 - Black Box Combinatorial Optimization with Monotone Structure: Applications in the IT Services Industry Nam Ho-Nguyen, Carnegie Mellon University, Tepper School of Business, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States, Giacomo Nannicini, Aly Megahed, Ahmed Nazeem We consider the NP-hard problem of optimizing a black box function over binary vectors subject to linear constraints, where the black box function satisfies a monotonicity property. We discuss two potential applications for such a problem: automated solution design of IT services contracts, and product selection for retail bundles with uncertain cost. We develop an iterative framework for choosing new binary vectors to query via monotonic polynomial regression, and give a valid optimality criterion for stopping. To scale our approach, we discuss potential heuristics for different steps of our framework. 1 - Blockchain Technology in Supply Chains: Recent Developments and Future Research

n WC12 North Bldg 126A Marketing II Contributed Session Chair: Mehmet Onal, Isik University, Ozyegin Unversity, Istanbul, 34794, Turkey 1 - An Analytical Model of Tier-based Loyalty Programs Amir Gandomi, Assistant Professor, Hofstra University, 031 Weller Hall, Hempstead, NY, 11549-1340, United States, Mehdi Nourinejad Loyalty programs are now a prominent market marketing strategy in many industries. They offer users a reward for becoming repeat consumers of a product. A prominent reward structure is the tier-based program that categorizes users according to their consumption intensity. Service providers like to design the tiers to maximize their expected profit. While research in this area is abundant, only a few provide analytical intuition of how the triers should be set up. This study The distinction between attributes and benefits is at the heart of product development. Some attributes are benefit enabling in the sense that their presence signals that an offering is responsive to a benefit being sought, while their absence indicates that the product does not provide the benefit. Not all brands, for example, are seen as luxurious and specific brand names may be needed for consumers to consider an offering as providing luxury. Benefit formation hinges on these types of attributes being present, while other attributes only enhance a benefit that has already been formed. We develop a model to identify attributes that are critical to a benefit formation and apply it to conjoint datasets. 3 - The Influence of Consumers’ Regulatory Focus and Cultural Dimensions on Intentions to Purchase from Online Stores Qiuling Li, University of Macau, Room 1048 E22, FBA, Universtity of Macau, Macau, Macao, Chanthika Pornpitakpan Consumer behavior in a multichannel context has increasingly received attention from academics. More and more people nowadays pursue online shopping. However, little research has examined how consumers’ motivations and cultural dimensions affect their shopping channel preferences. Addressing this gap of knowledge, this research examines how consumers’ regulatory focus (promotion focus vs. prevention focus) and cultural dimensions (i.e., individualism- collectivism, power distance, uncertainty avoidance, and masculinity-femininity, as identified by Hofstede) influence intentions to purchase from online stores versus physical stores. Several hypotheses are proposed. 4 - A Price Markdown Analysis for Two Vertically Differentiated Items Mehmet Onal, Ozyegin University, Istanbul, Turkey, Erhun Kundakcioglu We consider a retail store that sells two vertically differentiated products that can substitute each other. We assume that the two products differ in quality and price, and target two different customer segments. We also assume that items lose their utility linearly with time and may require price adjustments so that cannibalization does not occur, i.e., consumers from one segment never buy items targeted for the other segment. We identify optimal solution properties and propose an efficient solution approach. n WC13 North Bldg 126B Practice- Supply Chain Optimization II Contributed Session Chair: Jing Luo, University of Pittsburgh, Pittsburgh, PA, 15260, United States 1 - Supply Chain Design with Multi-modal Shipping and Varying Lead Times Gang Wang, University of Massachusetts Dartmouth, New Bedford, MA, 02740, United States This paper considers a multi-echelon supply chain design considering production scheduling, varying lead times, and multiple shipping options. The problem is to determine three types of decisions: a) strategic decision-the location of processing centers; b) tactical decision-the shipping quantities from suppliers to processing centers; and c) operational decisions-order assignment of demand points to processing centers as well as production schedules at both the suppliers and processing centers. We study three scenarios: 1) commit to delivery; 2) nonlinear shipping costs; and 3) complete delivery. finds the optimal tier design of loyalty programs. 2 - Benefit Formation and Enhancement Hyowon Kim, Doctoral Student, Ohio State University, 540A Fisher Hall 2100 Neil Ave, Columbus, OH, 43210, United States, Dong Soo Kim, Greg M. Allenby

473