Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

SA28

n SA25 North Bldg 131C

n SA28 North Bldg 221A Joint Session MSOM/APS: Supply Chain Management: Pricing, Shipping, and Expansion Sponsored: Manufacturing & Service Oper Mgmt Sponsored Session Chair: Nan Yang, University of Miami, Coral Gables, FL, 33146, United States Co-Chair: Panos Kouvelis, Washington University in St. Louis, One Brookings Drive, Campus Box 1156, Saint Louis, MO, 63130-4899, United States 1 - Shipping Consolidation Across Two Warehouses with Delivery Deadline and Expedited Options for E-commerce and Omni-channel Retailers Lai Wei, University of Michigan, Ann Arbor, MI, United States, Stefanus Jasin, Roman Kapuscinski Shipment consolidation is commonly used to avoid some of the shipping costs. However, when pending current orders are consolidated with future orders it may require more expensive expedited shipment to meet shorter deadlines. In this paper, we study the optimal consolidation policy focusing on the trade-off between economies of scale and expedited shipping cost. The optimal policies and their structures are characterized, where the impact of expedited shipment on both shipping policy and order fulfillment policy are explored. Two easily implementable heuristics are proposed, which perform within 1-2% of the optimal in intensive numerical tests. 2 - Joint Pricing and Inventory Optimization Under Minimax Regret Chengzhang Li, Purdue University, West Lafayette, IN, United States, Qi Feng, Mengshi Lu We study a robust pricing newsvendor problem under the min-max regret framework where the firm only knows the interval in which the random factor lies with high confidence in the demand model. We characterize the optimal decisions with a general demand model and further study the impact of inventory risk by comparing the optimal price and the risk-free price. We also study comparative statics of the optimal price with different unit cost and degree of demand ambiguity. Finally, our approach outperforms the traditional one in a data-driven setting with small sample sizes, high demand variability, or demand model misspecification. 3 - Supply Chain Competition: A Market Game Approach C Gizem Korpeoglu, University College London, Gower Street, London, WC1E 6BT, United Kingdom, Ersin Korpeoglu, Soo-Haeng Cho We study expansion and integration of supply chains where multiple suppliers sell to multiple retailers through a wholesale market. While the prior literature fails to capture retailers’ influence on the wholesale price, we develop a novel model based on a market-game mechanism that captures it. We show that more retailers raise the wholesale price and improve retailer profits, and hence the supply chain expansion is more beneficial than the prior literature finds. We then analyze integration of local supply chains, and find that this integration raises the total profit of all firms, but may reduce the total profit of firms in a retailer- oriented local supply chain with more retailers than suppliers. 4 - Impact of Price Postponement and Risk Aversion Under Supply Random Yield Panos Kouvelis, Washington University in St. Louis, One Brookings Drive, Campus Box 1156, Saint Louis, MO, 63130- 4899, United States, Guang Xiao, Nan Yang In this paper, we investigate the impact of price postponement and risk aversion under supply yield risk. More specifically, we study a risk averse monopoly firm’s production and pricing decisions under supply random yield with two distinct pricing schemes: ex ante pricing and responsive pricing. We adopt the conditional-value-at-risk (CVaR) as the risk aversion measurement, and investigate the impact of the firm’s risk aversion level on its optimal decisions and the corresponding profit.

Best Cluster Paper – I Sponsored: Service Science Sponsored Session Chair: Paul R. Messinger, University of Alberta, Edmonton, AB, T6G 2R6, Canada 1 - Service Science Best Cluster Paper Paul R. Messinger, University of Alberta, University of Alberta, Edmonton, AB, T6G 2R6, Canada This session consists of finalists presentations (judged by an expert panel) to determine the Service Science Section Best Cluster Paper. n SA26 North Bldg 132A Approximations and Controls of Queues Sponsored: Applied Probability Sponsored Session Chair: Yunan Liu, NCSU 1 - Staffing and Scheduling to Differentiate Service Levels in Multiclass Overloaded Queues with Time-varying Arrivals Xu Sun, Columbia University, Kyle Hovey, Yunan Liu Motivated by large-scale service systems (e.g., emergency departments), we study an overloaded multi-class queueing system having time-varying arrivals. Our objective is to devise appropriate staffing and scheduling policies to achieve differentiated service levels for each customer class. Our proposed policy is both time dependent (coping with the time variability in arrival pattern) and state dependent (capturing the stochastic variability in service times and arrival times). Effectiveness of our staffing and scheduling rules are confirmed by heavy traffic limit theorems (with the system scale increases) and computer simulation experiments. 2 - Many Server Scaling of the N-system Under FCFS-LISF Dongyuan Zhan, University College London, Gideon Weiss The N-System with independent Poisson arrivals and exponential server- dependent service times under first come first served and assign to longest idle server policy has explicit steady state distribution. We scale the arrival and the number of servers simultaneously, and obtain the fluid and central limit approximation for the steady state. This is the first step towards exploring the many server scaling limit behavior of general parallel service systems. 3 - On the Stability of Large-scale Markovian Parallel Server Networks Guodong Pang, PhD, Penn State University We will discuss stability of Markov parallel server networks with/without abandonment in the Halfin-Whitt regime. In particular, exponential ergodicity properties are presented for both the limiting controlled diffusions and the diffusion-scaled queueing processes under various scheduling policies. 4 - State Dependent Pricing in Naor Model with Arrival Rate Uncertainty Chengcheng Liu, The University of Texas at Austin, Austin, TX, United States, John Hasenbein This paper examines extensions of Naor’s observable queueing model in which arrival rate is not known with certainty by either customers or system managers. We consider the arrival rate as a non-degenerate random variable with unknown realizations. This work is also an extension of the study of Chen and Hasenbein on a related model in that their model only considers homogeneous customer populations. In view of a social optimizer (SO) and a revenue maximizer (RM), we analyze the social benefit rate and revenue rate, and investigate optimal state- dependent pricing policies in the presence of arrival rate uncertainty. 5 - On Brownian Approximation for Superposition of Renewal Processes Shuangchi He, National University of Singapore, 1 Engineering Drive 2, Dept. of Industrial and Systems Engineering, Singapore, 117576, Singapore We investigate the superposition of many stationary renewal processes. A centered and scaled version of this superposition process is known to weakly converge to a Gaussian process, which is in general not a Brownian motion. By an expansion of the covariance function, we discuss when the superposition process can be accurately approximated by a Brownian motion. We also discuss an application to many-server queues with a general service time distribution in the efficiency-driven regime.

11