2015 Informs Annual Meeting

MC03

INFORMS Philadelphia – 2015

2 - Stochastic Network Interdiction with Risk Preference Jing Zhang, University at Buffalo, SUNY, 338 Bell Hall, Buffalo, NY, 14221, United States of America, jzhang42@buffalo.edu, Jun Zhuang, Brandon Behlendorf This paper studies the stochastic network interdiction problem, where the defender maximizes the length of the shortest path between a source and a destination by allocating sensors to the arcs with a limited budget. There is a detecting probability of the sensor, and the defender is unaware of the type of the attacker (strategic, or non-strategic). We develop game-theoretic models, solution methods, and illustrate the models using a portion of the Arizona-Mexico border transportation network. 3 - Keeping Pace with Criminals: Designing Patrol Allocation Against Adaptive Opportunistic Criminals Milind Tambe, USC, 941 Bloom Walk, Los Angeles, CA, United States of America, tambe@usc.edu, Arunesh Sinha, Chao Zhang A distinctive feature of urban crimes is that criminals react opportunistically to patrol officers’ assignments. Opportunistic criminals are less strategic in planning attacks and flexible in executing them. Our goal is to recommend optimal police patrolling strategy against such opportunistic criminals. Our key contribution is to learn the criminal model from real-world crime and patrol data by representing the criminal behavior as parameters of a Dynamic Bayesian Network. 4 - Improving Logistics Security by using Distributed Container Inspection History Data Gary Gaukler, Drucker School of Management, Claremont Graduate University, Claremont, CA, 91711, United States of America, Gary.Gaukler@cgu.edu We present a two-stage interdiction model for smuggled nuclear materials in which prior container inspection data from an upstream inspection stage is used as a low-cost way of increasing overall interdiction performance. We provide insights into how a decision maker at a downstream inspection stage should optimally use detection data from the upstream stage to improve the overall detection capability. Innovative Scheduling Applications Cluster: Scheduling and Project Management Invited Session Chair: Tolga Aydinliyim, Baruch College, One Bernard Baruch Way, Dept of Management Box B9-240, New York, NY, United States of America, Tolga.Aydinliyim@baruch.cuny.edu 1 - Throughput Optimization in Single and Dual-gripper Robotic Cells Manoj Vanajakumari, Texas A&M University, 3367 TAMU, College Station, TX, 77845, United States of America, manojuv@tamu.edu, Chelliah Sriskandarajah, Sushil Gupta In view of maximizing throughput, practitioners uses a class of cycles known as 1- unit cycles in which the cell returns to the same state after the production of each unit. The complexity of throughput optimization in the class of 1-unit cycles in single and dual-gripper robotic cells is the main focus of this paper. We provide some insights for throughput optimization using two-unit cycles. 2 - A Decision Support System for Appointment System Templates with Operational Performance Targets William Millhiser, Associate Professor, Baruch College, One Bernard Baruch Way, Box B9-240, New York, NY, 10011, United States of America, William.Millhiser@baruch.cuny.edu, Emre Veral We present a web-based scheduling system for outpatient services that meets user-defined operational targets to achieve managed/fair waiting times, dependable session end times, and minimal unintended idle time for providers. Using historical service times and an underlying model based on prior research, we demonstrate that appointments that meet these operational targets can be scheduled in a real-time environment, while the software provides dynamic assistance in selecting appointment slots. 3 - Improving Blood Products Supply through Donation Tailoring Ali Ekici, Assistant Professor, Ozyegin University, Industrial Engineering, Nisantepe Mah., Orman Sok, Cekmekoy, Istanbul, 34794, Turkey, ali.ekici@ozyegin.edu.tr, Elvin Coban, Okan Orsan Ozener Multicomponent apheresis (MCA) allows the donation of more than one component and/or more than one transfusable unit of each component. It provides several opportunities including (i) increasing the donor utilization, and (ii) tailoring the donations based on demand. In this study, we develop mathematical models to develop donation schedules for repeat donors while considering factors such as blood products demand, shelf-life of the blood products, donation costs, and deferral times. MC03 03-Room 303, Marriott

4 - Optimal Schedule of Elective Surgery Operations Subject to Disruptions by Emergencies Xiaoqiang Cai, The Chinese Unievrsity of Hong Kong, Shatin, Hong Kong, Hong Kong - PRC, xqcai@se.cuhk.edu.hk, Xianyi Wu, Xian Zhou Elective surgery operations are to be scheduled at an operating theater, which can accommodate one operation at a time. Emergency cases may arrive randomly, which have higher priority. Any operation, no matter it is normal or emergent, has to be processed until it is completed. Optimal dynamic policies are derived. 5 - Optimal Movement and Transshipment of Rail Freight Shipments Chinmoy Mohapatra, PhD Candidate, University of Texas at Austin, 3500 Greystone Drive, Apt. 126, Austin, TX, 78731, United States of America, chinmoym@utexas.edu, Anant Balakrishnan We study the problem of assigning shipments to scheduled transport services that share common capacitated resources. At each node, shipments using same outbound service are assigned in a first-in first-out order. We develop modeling and algorithmic enhancements to effectively solve this large-scale optimization problem, and present computational results for real-life instances. MC04 04-Room 304, Marriott Joint Session JFIG/MIF: Panel Discussion on Tenure and Promotion Sponsor: Junior Faculty Interest Group Sponsored Session Chair: Shengfan Zhang, Assistant Professor, University of Arkansas, 4207 Bell Engineering Center, Fayetteville, United States of America, shengfan@uark.edu Co-Chair: Lauren Davis, North Carolina A&T State University, 1601 E. Market St., Greensboro, NC, United States of America, lbdavis@ncat.edu 1 - Department Chair Panel Moderator: Shengfan Zhang, Assistant Professor, University of Arkansas, 4207 Bell Engineering Center, Fayetteville, United States of America, shengfan@uark.edu, Panelists: Mark Daskin, Scott Grasman, Ann Marucheck, Alice Smith A session with IE and business department chairs on issues related to junior faculty. MC05 05-Room 305, Marriott Predictive Models of Human Behavior in Social Media Cluster: Social Media Analytics Invited Session Chair: Tauhid Zaman, MIT Sloan School of Management, 50 Memorial Drive, Cambridge, MA, 02139, United States of America, zlisto@mit.edu 1 - Adaptive Searches in Twitter Chris Marks, MIT, 50 Memorial Drive, Cambridge, MA, 02139, United States of America, cemarks@mit.edu We present a methodology for adaptively collecting data from the Twitter microblogging application. Based on an initial search query or filter, our method uses network structure and count data from the returned results to update the search query so that additional relevant results are returned. Measures of result relevance will also be presented and discussed. 2 - Graph Control over Social Media: The Follow-back Problem Krishnan Rajagopalan, Graduate Student, MIT, 50 Memorial Drive, Cambridge, MA, 02139, United States of America, krishraj@mit.edu, Tauhid Zaman We create a new influence maximization problem on social media where an agent seeks to form a connection with a specific user, the target, in an online social network. We model the problem as an MDP. We use transition probabilities, learned from analysis of Twitter data and find a policy that gives the agent the optimal sequence of interactions with the target’s friends to maximize the probability the target will form a connection with the agent. We identify heuristics for certain topologies.

206