2016 INFORMS Annual Meeting Program

WA89

INFORMS Nashville – 2016

2 - The Expanding Search Ratio Of A Graph Thomas Lidbetter, London School of Economics, Houghton Street, London, WC2A 2AE, United Kingdom, t.r.lidbetter@lse.ac.uk Thomas Lidbetter, Rutgers Business School, Newark, NJ, 07102, United States, t.r.lidbetter@lse.ac.uk, Spyros Angelopoulos, Christoph Dürr We study the problem of searching for a hidden target in an edge-weighted graph. We use a recently introduced search paradigm called expanding search, where a sequence of edges is chosen starting from a given root vertex such that each edge is adjacent to a previously chosen edge. We define the search ratio of an expanding search as the max. over all vertices of the ratio of the time taken to reach the vertex and the shortest-path cost to it from the root. We seek the randomized expanding search with minimum expected search ratio, which is equivalent to solving a zero-sum game between a Searcher and a Hider. We solve the problem for certain classes of graphs, and obtain constant-factor approximations for others. 3 - Modeling Short Range Missile Defense And Iron Dome Michael J Armstrong, Associate Professor, Brock University, Dept of FOIS Faculty of Business, St Catharines, ON, L2S 3A1, Canada, michael.armstrong@brocku.ca This paper develops a model of short range ballistic missile defense and uses it to study the performance of the Israeli Iron Dome system. The deterministic model allows for inaccurate missiles, unsuccessful interceptions, and civil defense. Model enhancements consider the trade-offs in attacking the interception system, the difficulties faced by militants in assembling large salvos, and the effects of imperfect missile classification by the defender. A stochastic model is also developed. Analysis shows that system performance is highly sensitive to the missile salvo size. 4 - United States Coast Guard Reduced Staffing Simulation Chad A Long, United States Coast Guard, FAA Technical Center Building 350, Atlantic City International Airport, Atlantic City, NJ, 08234, United States, chad.a.long@uscg.mil Under the current schedule, Air Station Atlantic City pilots are averaging 17 flight hours per month, which is below the United States Coast Guard requirement. A new schedule, with decreased staffing was simulated in Arena, to determine if it would increase flight time and be robust enough to staff the unit year-round. 5 - Behavioral Analysis Of Illegal Fishery In The Gulf Of Mexico Ali Pala, PhD Student, University at Buffalo, SUNY, 271 Palmdale Drive, Apt 5, Buffalo, NY, 14221, United States, alipala@buffalo.edu, Vineet Madasseri Payyappalli, Jun Zhuang Illegal fishery activities in the Gulf of Mexico pose a threat to U.S. national security, as well as damage to the economy. The U.S. Coast Guard estimates over 1,100 incursions annually by Mexican fisherman into U.S. regulated waters in the Gulf of Mexico. In this research, we study the behavior of illegal fishermen and analyze how that is affected by various factors such as weather, fish habitats, seasonal effects, and others. We use data mining techniques and mathematical models to propose optimal patrolling strategies that improve the chances of interdicting illegal fishery. WA89 Broadway C-Omni Innovations in Connected and Autonomous Vehicles Sponsored: TSL, Intelligent Transportation Systems (ITS) Sponsored Session Chair: Peiheng Li, Arizona State University, 1, Tempe, AZ, 6, United States, peihengl@asu.edu 1 - A Distributed Framework For Network-Wide Traffic Monitoring And Platoon Information Aggregation Using V2V Communications Peiheng Li, Arizona State University, Tempe, AZ, United States, peihengl@asu.edu, Yingyan Lou This study explores an innovative framework for distributed traffic monitoring and information aggregation using vehicle-to-vehicle communications alone. Distributed protocols that run on individual vehicles are developed and tested through simulation. A new concept of micro-discontinuity is proposed for identifying platoons, and initiating and terminating traffic information aggregation. The impact of market penetration rate is also analyzed. 2 - Development Of Adaptive Forward Collision Warning Algorithm Incorporating Prediction Uncertainty To Reduce False Alarms Yong Hoon Kim, Purdue University, West Lafayette, IN, United States, kim523@purdue.edu, Shubham Agrawal, Srinivas Peeta This study proposes an adaptive forward collision warning (FCW) algorithm that adjusts its warning threshold according to the uncertainty in vehicle trajectory prediction. The prediction uncertainty is modeled by kinematic information of surrounding vehicles obtained through vehicle-to-vehicle communication. Numerical experiments show that the proposed adaptive FCW algorithm can provide early warning while reducing the rate of false/nuisance alarms.

3 - Integrating The Operations Perspective Into Planning Of Automotive Vehicle Platforms Paul Jana, Research assistant, Technische Universität Muenchen, Arcisstr 21, Munich, 80333, Germany, paul.jana@tum.de, Martin Grunow To address the mass customization trade-off many automotive OEMs moved towards a multi-platform based product portfolio. One major benefit is the reduced internal variety in operations. So far, the multi-platform problem is mostly approached from the product perspective without adopting the operations perspective. We present a mixed-integer linear programming approach integrating network design, capacity configuration, and further operations requirements into automotive multi-platform planning and show the superiority of an integrated approach in a numerical study based on an industry setting. WA90 Broadway D-Omni Health Care, Modeling XII Contributed Session Chair: Yang Zhan, Shanghai Jiao Tong University, Shanghai, 200030, China, zhanyangjy@sjtu.edu.cn 1 - A Stochastic Stackelberg-Nash-Cournot Equilibrium Model For Long-term Care Capacity Planning Ashkan Hassani, Texas A&M University, 4050 ETB, College Station, TX, 77840, United States, ashkanhassani@tamu.edu, Mark Alan Lawley, Nan Kong Long-term care (LTC) is needed by people who have limitations in activities of daily living. In general, there are two types of LTC delivery: nursing home care and home- and community-based service (HCBS). We develop a stochastic Stackelberg-Nash-Cournot (SNC) equilibrium model to determine the optimal capacities for nursing home and HCBS. The capacity of HCBS will be determined by the public insurer who performs as a leader in a Stackelberg game. Nursing homes will be followers in the Stackelberg game with knowledge of the public insurer’s policy, and they compete with each other and could reach Nash equilibrium in a Nash-Cournot game. 2 - Dynamically Tracking Multiple Types Of Risks Of Falls With Wearable Sensors via Data Association Problem Wearable sensors are used in practice for detection of falls. We formulate the data association problem arising in dynamic tracking multiple types of risks of falls with wearable sensors as the multi-dimensional problem (MAP). Tracking of risks is done dynamically as data from multiple sensors arrive continually and are partitioned into frames. We consider multiple-frame risk-track initiation and risk- track maintenance for single-pane, double-pane, and multiple-pane sliding windows. Track maintenance include: extending existing risk-tracks, terminating existing tracks, and initiating new risk-tracks. The optimization models are applied to empirical data from wearable sensor system. 3 - A Simulation Model Of Glucose Metabolism For Predicting Blood Glucose Level Buket Aydas, University of Wisconsin-Milwaukee, Milwaukee, WI, 53202, United States, baydas@uwm.edu, Mukul Goyal We develop a human body glucose metabolism model for predicting blood glucose levels using discrete event simulation. The model estimates the glycated hemoglobin (HbA1c) level in the blood and predicts the impact, in terms of BGL and HbA1c levels, of a particular diet-exercise routine. Each organ in the human body that relates to glucose metabolism is modeled individually within the simulation. The results are validated by using reference glycemic index (GI) values of food. 4 - Two-level Optimization To Balance Value Flow And Patient Flow In Operating Room Block Scheduling Wei Li, Assistant Professor, University of Kentucky, 414J CRMS Building, 147 Graham Avenue, Lexington, KY, 40506-0108, United States, wei.mike.li@uky.edu Operating room (OR) block scheduling in the long term is important for OR scheduling in the medium term and OR control in the short term. Similar to the trade-off between production cost and holding cost in manufacturing systems, there is a trade-off between value flow and patient flow in OR scheduling. We use a two-level optimization model to balance the trade-off in OR scheduling, and test the robustness of the model through simulation. Alla Kammerdiner, Assistant Professor, New Mexico State University, Las Cruces, NM, United States, alla@nmsu.edu, Andre Nunes Guerrero

394