Informs Annual Meeting 2017

TB21

INFORMS Houston – 2017

3 - Consumption Decisions in Queues Sezer Ulku, Georgetown University McDonough School of Business, 594 Rafik B. Hariri Building, Washington, DC, 20057, United States, su8@georgetown.edu, Christopher Hydock, Shiliang Cui We study how the time spent waiting in line influences consumption decisions. We conduct a series of lab experiments and field studies in a variety of contexts, and observe that people deviate from rational behavior; we find that the longer people wait in line, the larger is their purchase quantity. In our final experiment, we show that two strategies that are used to improve consumer experience in queues can in fact result in lower purchase quantities. 4 - When Should Customers Control Service Delivery? Implications for Service Design Ioannis Bellos, George Mason University, Enterprise Hall, 4400 University Dr., MS.5F4, Fairfax, VA, 22030, United States, ibellos@gmu.edu, Stylianos Kavadias What do a Mongolian stir-fry restaurant, a discount broker, and a lab specializing in home-test kits have in common? They are examples of innovative services that base their success on customers’ participation in the service delivery. Motivated by these practices we develop an analytical model to determine which parts of a service process should be performed by the service provider and which parts should be delegated to the customers. 342C Supply Chain Risk and Disruptions Invited: InvitedNatural Hazard Planning Invited Session Chair: Eghbal Rashidi, Clemson University, erashid@clemson.edu 1 - Supply Chain Risk Analysis using Dynamic Fault Tree Xue Lei, Iowa State University, Ames, IA, 50011, United States, xlei@iastate.edu Supply chain risk analysis can help supply chain managers assess the risk in their supply chain and identify mitigation strategies. This research applies dynamic fault trees to assess the risk in manufacturing supply chains. Dynamic fault trees were developed to measure the reliability of systems in which the timing of failure for different components impacts total system failure. Applying dynamic fault trees to quantify the risk in a supply chain allows us to analyze how the time at which a primary supplier fails impacts the ability of a backup supplier to meet the delivery schedule. A new dynamic gate is proposed to consider the situation in which a firm relies on two primary suppliers. 2 - Risk Mitigation Models for Low Volume High Value (LVHV) Supply Chains Michael Daniel Sherwin, Mississippi State University, 1767 Independence Way, Valencia, PA, 16059, United States, mdsherwin@gmail.com, Hugh Medal Firms that produce low volume high value (LVHV) goods are more susceptible in many ways to delays within their respective supply chains than higher volume, higher value manufacturers. Due to the scarcity of capable suppliers to produce large, complex, highly precise items combined with the cost to develop a new supplier, a large number of LVHV firms have many single or sole source suppliers within their supply chains. We define supplier unreliability as the probability that a given product or service is not delivered on-time and develop a multinomial logistic regression model based on the variables that affect the unreliability of a LVHV supplier. 3 - Simulating Severe Supply Chain Disruptions with Multiple Suppliers and Firms This presentation discusses a model and simulation of a severe supply chain disruption in which several suppliers suffer from disabled production facilities and firms that purchase goods from those suppliers may suffer a supply shortage. The simulation is applied to a case inspired by the Japanese tsunami. 5 - A Multi Stage Stochastic Programming Model for Semiconductor Supply Chain Risk Management Eghbal Rashidi, Clemson University, Clemson, SC, United States, erashid@clemson.edu We propose a three-stage stochastic programming model for a multi-product semiconductor supply chain risk management problem with uncertainty in process yield and product demand. The production processes are considered to follow a push policy in the Front-End, and a pull policy in the Back-End; with the die bank being the decoupling point. Outsourcing of production and process is TB21 Cameron MacKenzie, Iowa State University, 3029 Black Engineering, Industrial and Manufacturing Systems Eng, Ames, IA, 50011, United States, camacken@iastate.edu

considered for each process step in this study, at fabrication, sort, assembly and test. Computational results on test case problems show that our proposed stochastic programming model produces solutions that are significantly better than those produced by a deterministic expected value approach. 4 - Effectiveness of Resilience Strategies Against Supply Chain Disruptions: An Empirical Study Milad Baghersad, Virginia Polytechnic Institute and State, Pamplin Hall, Room 1007, Blacksburg, VA, 24061, United States, mbaghers@vt.edu, Christopher Zobel In this study, we empirically evaluate the effectiveness of three specific resilience strategies- operational slack, business diversification, and geographical diversification- in reducing the negative impacts of supply chain disruptions on firms’ financial performance. Borrowing from the systems resilience literature, we develop four complementary metrics to quantify the negative impacts of supply chain disruptions on firms’ performance. By adopting a more comprehensive view of firms’ performance, we are able to develop new and expanded inferences about the types of impacts that the different strategies have on firms’ performance after supply chain disruptions. 342D Multi-product Assortment, Pricing and Inventory Control Sponsored: Revenue Management & Pricing Sponsored Session Chair: Anran Li, Columbia University, New York, NY, United States, al2942@columbia.edu 1 - Complementarity Analysis for Optimal Multi Item Inventory and Partial Price Management TB22 We study a multi-product joint inventory replenishment and pricing problem, where some pairs of products are economic complements but other pairs are substitutes. The seller periodically replenishes the products and controls the selling price of one of them while the other prices are exogenously determined. We show that there always exists an economic relationship between products that the seller should reflect in the optimal operations, and moreover, such a relationship may not be identical to that from the customers’ viewpoint. Further, the optimal decisions should follow a base-stock list-price policy in which there Sajjad Najafi, University of Michigan, Ann Arbor, MI, United States, snajafi@umich.edu, Sami Najafi Asadolahi, Chi-Guhn Lee, Steven Nahmias We study a firm offering a line of perishable products with fixed initial inventory over a finite horizon. Consumers arrive randomly and inspect products sequentially until they find a product to purchase (if any). Each consumer incurs a positive cost to inspect a product and hence may stop the sequential search without inspecting all the available items. The firm aims at offering the best sequence of products, as well as the best price for each product at each time. We show that, under some circumstances, (i) it is optimal to present products in a descending order of quality; (ii) surprisingly, the optimal price can increase over time. 3 - Approximation Algorithms for Product Framing and Pricing Anran Li, Columbia University, Mudd 345, Columbia University, New York, NY, 10027, United States, al2942@columbia.edu, Guillermo Gallego, Van-Anh Truong, Xinshang Wang We propose one of the first models of “product framing” and pricing. Product framing refers to the way consumer choice is influenced by how the products are framed, or displayed. We present a model where a set of products are displayed into a set of virtual web pages. We assume that consumers consider only products in the top pages, with different consumers willing to see different numbers of pages. Consumers select a product, if any, from these pages following a general choice model. We show that the product framing problem is NP-hard. We derive algorithms with guaranteed performance relative to an optimal algorithm under reasonable assumptions. We also present structural results for the pricing problem under framing effects. exist a set of ideal values of the inventory levels and the price. 2 - Dynamic Pricing under Consumer’s Sequential Search Sang Jo Kim, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong, sangjo@ust.hk, Youyi Feng, Jianjun Xu

310