Informs Annual Meeting 2017

SB16

INFORMS Houston – 2017

SB14

the nature of incentives. Driver effort on the platform is unobserved, for which we devise a novel local matching model based imputation method. We find that in short-term (current week) passenger incentives are more effective while the opposite is true in the long-term (next 3 months). 2 - (DIS)Value of Information in a Supply Chain of Perishable Products Ming Hu, University of Toronto, Rotman School of Management, Our study is based on a data set from a fast-food restaurant chain. During the observation window, this restaurant chain switched from third-party POS systems to their own system, which allows them to gather real-time sales information from stores. We study store managers’ adjustment of operational decisions in response to this change, as well as the resulting impact on customer satisfactions. 3 - Behavioral Drivers of Routing Decisions: Evidence from Restaurant Table Assignment Fangyun Tan, Southern Methodist University, 6212 Bishop Blvd, Dallas, TX, 75275, United States, ttan@cox.smu.edu, Bradley R. Staats In many settings, humans make routing decisions dynamically, either because algorithms don’t exist, decision support tools have not been implemented, or existing rules are not enforced. Understanding how individuals make decisions creates the opportunity to identify both positive deviances, as well as suboptimal decision making that can be improved. In this paper we theoretically identify the factors that may impact decision making before empirically examining a large operational data set in a casual restaurant setting to research whether and how hosts deviate from their predefined round-robin rule to seat customers to servers. 4 - Empirical Analysis of Supply Network Relationship Concentration, Structure, and Firm Performance Marcus A. Bellamy, Boston University, Rafik B. Hariri Building, 595 Commonwealth Avenue, Boston, MA, 02215, United States, bellamym@bu.edu, Soumen Ghosh, Manpreet Singh Hora We examine the concentration and structural configuration of a firm’s supply chain as drivers of its performance using supply chain relationship data from the Bloomberg database. We demonstrate how firm performance may be influenced by such drivers and how the industry context within which a firm operates augments their influence. 332F Inventory Management Contributed Session Chair: Mustafa Hekimoglu, Isik University, Istanbul, Turkey, mustafa.hekimoglu@isikun.edu.tr 1 - Joint Replenishment and Transshipment for Three Locations - Asymptotics and Bounds Weifen Zhuang, Associate Professor, School of Management, Xiamen University, Xiamen, 361005, China, wfzhuang@xmu.edu.cn We study the problem of joint replenishment and transshipment for a retailer who sells seasonal products through three physical stores. The decisions involve the one-shot stocking at the beginning of the season and the supply/transshipment decision throughout the season. Applying an MDP formulation with compound Poisson demand processes, we identify the optimal policies. We then study two downward transshipment models and develop bounds, which are shown to be asymptotically optimal. We develop effective heuristics by making use of the bound solution. The bounds and heuristics can be extended to deal with the problem of multiple-location. 2 - Ordering and Transshipment Policies for a Healthcare System with Two Hospitals Erhun Kundakcioglu, Ozyegin University, Faculty of Engineering, Nisantepe mah. Orman sok., Istanbul, 34794, Turkey, erhun@alumni.ufl.edu, Hamed Shourabizadeh One of the most important challenges in healthcare systems is managing inventory. Depending on available capacity and average expiration dates, hospitals typically determine replenishment cycles without mathematical optimization. Our study focuses on determining optimal inventory levels that minimize the sum of inventory holding, ordering, and transshipment costs for a system of two hospitals. Hospitals can collaborate in the form of stock transfer and joint replenishment. Due to uncertain demand, hospitals are expected to benefit from a collaborative inventory model that increases the quality of care for patients (drug accessibility) and time between consecutive orders. 105 St. George Street, Toronto, ON, M5S.3E6, Canada, ming.hu@rotman.utoronto.ca, Peng Luo, Jiahua Wu SB16

332D Stochastic Modeling in Healthcare Operations Sponsored: Manufacturing & Service Oper Mgmt, Healthcare Operations Sponsored Session Chair: Carri Chan, Columbia Business School, New York, NY, 10027, United States, cwchan@columbia.edu Co-Chair: Vahid Sarhangian, Columbia University, vs2573@columbia.edu 1 - Managing Patient Flow Dynamics in Internal Wards Jing Dong, Northwestern University, 2145 Sheridan Road, Room C210, Evanston, IL, 60208, United States, jing.dong@northwestern.edu, Ohad Perry We introduce a new class of queueing models that take into account the most salient features of patient flow dynamics in Internal Wards. We also develop efficient approximation techniques for long-run performance analysis, which greatly facilitate evaluating the effect of various operational policies. 2 - Data-driven Appointment-scheduling under Uncertainty: The Case of an Infusion Unit in a Cancer Center Nikolaos Trichakis, MIT, 100 Main Street, E62-576, Cambridge, MA, 02143, United States, ntrichakis@mit.edu, Avishai Mandelbaum, Petar Momcilovic, Sarah Kadish, Ryan Leib, Craig Bunnell We consider appointment scheduling under a time-varying number of servers, in a data-rich environment where service durations and punctuality are uncertain. Our data-driven approach, based on infinite-server queues, yields tractable and scalable solutions that accommodate hundreds of jobs and servers. We successfully test our approach against near-optimal algorithms, which exist for merely single-servers. To test for practical performance, we leverage a unique dataset from a cancer center, that combines real-time locations, electronic health records and appointments log. Focusing on one of the center’s infusion units (~90 Rouba Ibrahim, University College London, MS& I.department, UCL, Gower Street, London, WC1E 6BT, United Kingdom, rouba.ibrahim@ucl.ac.uk, Vedat Verter, Beste Kucukyazici, Michel Gendreau, Mark Blostein We develop an analytical framework for personalizing the anticoagulation therapy of patients who are taking warfarin, and present results from a case study using data collected at the anticoagulation clinic of the Jewish General Hospital in Montreal. 4 - Dynamic Server Assignment in a Multiclass Queueing System with Shifts, with Application to Nurse Staffing in Emergency Departments Vahid Sarhangian, University of Toronto, Toronto, ON, Canada, sarhangian@mie.utoronto.ca, Carri Chan Nurse staffing decisions in emergency departments (EDs) are typically assigned weeks in advance, which can create staffing imbalances as patient demand fluctuates. In this work, we consider the potential benefits of assigning nurses to different areas within an ED at the beginning of each shift. We study the problem of optimal reassignment of nurses to areas by considering a multiclass queueing model of the system. We analyze an associated fluid control problem and use the solution to develop policies that achieve asymptotically optimal performance under fluid-scaling for the original stochastic system. We find this additional flexibility can substantially reduce waiting times for patients. 332E Empirical Supply Chain Research Sponsored: Manufacturing & Service Oper Mgmt, Supply Chain Sponsored Session Chair: Nitish Jain, London Business School, London, NW1 4SA, United Kingdom, njain@london.edu 1 - Estimating Consumer Preferences for Pooled Transportation Kashish Arora, INSEAD, Fontaniebleau, France, kashish.arora@insead.edu, Karan Girotra, Fanyin Zheng Marketplace operators run aggressive incentive schemes to achieve scale, that is key to the efficacy, survival and eventual domination of a marketplace. This study quantifies and compares the effect of incentives given to the “buyer” side and “seller” side using data from a leading ride-hailing market. We build a structural model to accurately capture the driver and passenger response to incentives, and daily appointments, 25+ infusion chairs), we reduce cost by 30%. 3 - Designing Personalized Anticoagulation Therapy SB15

39