Informs Annual Meeting 2017

SC56

INFORMS Houston – 2017

SC56

Performance (OTP). We develop a simulation-based approach to measure the operability of a given schedule under various uncertainties. We aim to optimize the schedules by allocating the available time among the turn and block times so that OTP is maximized at minimal cost using a mixed integer programming formulation. 4 - Simulationof Collision Risk for Alternate Airspace Architectures John Shortle, George Mason University, Dept of Systems Engineering & OR, 4400 University Drive MS.4A6, Fairfax, VA, 22030-4444, United States, jshortle@gmu.edu Future airspace architectures will need to handle a wide diversity of aircraft operating in a semi-autonomous fashion. This talks presents a simulation of an airspace architecture in which the core requirement is maintaining a target level of collision risk in all regions of the airspace. Aircraft with better collision avoidance capabilities are thus able to fly in denser airspace. The talk presents a simulation of the concept and discusses implications to capacity, delay and collision risk. 362E Data Collection and Mining on Traffic Networks Sponsored: Transportation Science & Logistics Sponsored Session Chair: Seyedbehzad Aghdashi, North Carolina State University, 112 Gwinnett Pl, Cary, NC, 27518, United States, saghdas@ncsu.edu 1 - Travel Time Estimation in the Age of Big Data Arthur J. Delarue, Massachusetts Institute of Technology, 77 Massachusetts Ave, Cambridge, MA, 02139, United States, adelarue@mit.edu, Dimitris Bertsimas, Patrick Jaillet, Sebastien Martin In this talk, we show that we can leverage network optimization insights to tractably extract accurate travel time estimations from taxi travel time data, with the additional ability to take advantage of many other sources of traffic information. Using synthetic data, we establish the robustness of our algorithm to uncertainty, and display its ability to significantly reduce input variance. We show on real data that our algorithm can leverage any avail- able amount of data, even in a high-variance environment, to provide accurate insights about urban traffic patterns on different scales. 2 - Understanding City Dynamics from Taxis Trajectory Data Shirin Najafabadi, PhD Student and Research Assistant, The City University of New York (CCNY), 160 Convent Ave, Steinman Hall - Room T-117, New York, NY, 10031, United States, snajafabadi@ccny.cuny.edu, Mahdieh Allahviranloo This study focuses on trajectories obtained from GPS-enabled taxis and their applications for mining urban commuting patterns. A case study with a taxi trajectory data-set in New York, USA is presented to demonstrate and evaluate spatial distribution of taxi ridership and reveal facts about city dynamics. Using grid cell decomposition method, taxi data dis-aggregated into uniform size units and a radiation model was applied to model spatial correlations of factors associated with urban density to taxi ridership. Also, this paper presents a novel approach using temporal attractiveness of POIs (Point of Interests) to identify activity-locations as well as duration from raw GPS trajectory. 3 - An Optimization Approach to Install License Plate Recognition Sensors in a Traffic Network for Full Road Flow Monitoring Considering the Measurement Errors This work addresses the network sensor location problem (NSLP) to install the minimum number of license plate recognition (LPR) sensors on roads in a traffic network to meet full road flow monitoring. The objective function of the model also aims at minimizing the total measurement errors of sensors while meeting the budget constraint. The Monte Carlo simulation is then applied to consider the randomness of the measurement errors in finding the optimum location set of sensors. Eventually, the concept of backup sensors is developed to maintain the full road flow monitoring regarding the possibility of sensors failures in monitoring the flows. Mostafa Salari, Research Assistant, University of Calgary, 312 3420 50th St., NW, Calgary, AB, T3A 2E1, Canada, Mostafa.salari2@ucalgary.ca, Lina Kattan SC58

362C Continuous Facility Location Sponsored: Location Analysis Sponsored Session Chair: Atsuo Suzuki, PhD, Nanzan University, 18 Yamazato-cho, Nagoya, 466-8673, Japan, atsuo@nanzan-u.ac.jp 1 - Multi-range Sensor Location Problem Daisuke Watanabe, Tokyo University of Marine Science and Technology, 2-1-6 Etchujima, Koto-ku, Tokyo, 135-8533, Japan, daisuke@kaiyodai.ac.jp, Richard Church In this study, we formulate the optimal location model for surveillance platform with multi-range sensor based on the Maximum covering location model. The objective is to maximize the probability of detecting a historical set of piracy attacks. Using this model, we solve the optimal location of the Counter-Piracy Surveillance System in Somalia. 2 - The Big Triangle Small Triangle Method for the Weber Problem with Line Barriers Atsuo Suzuki, Nanzan University, Dept of Systems and Mathematical Sciences, 18 Yamazato-cho, Nagoya, 466-8673, Japan, atsuo@nanzan-u.ac.jp, Zvi Drezner, Tammy Drezner We apply the Big Triangle Small Triangle (BTST) method to the Weber problem with line barrier on the plane. We assume that there are demand points on the plane, and a line barrier with several passing points separates the demand points into two groups. It is a simple model of the situation where a river with bridges prohibits the direct access for the users to the facility on the other side of the river. The objective function of the problem is not convex because it is represented as the minimum of the convex functions. While a naive descent method fails to obtain the exact solution, the BTST algorithm obtains the exact solution. 362D Aviation Applications in Simulation Sponsored: Simulation Sponsored Session Chair: John Shortle, George Mason University, George Mason University, Fairfax, VA, 22030-4444, United States, jshortle@gmu.edu 1 - A Distance-based Network Simulation for Airport Runway and Priority Scheduling Haluk Damgacioglu, PhD Candidate, University of Miami, 1251 Memorial Drive McArthur Engr Bldg Room 280, Coral Gables, FL, 33146, United States, haluk.damgacioglu@miami.edu, Alyssa Guller, Nurcin Celik Runways are the main bottlenecks in airport ground networks leading to extended flight delays and hindering with its throughput. Here, we develop a distance-based network simulation for runway and flight scheduling to minimize the total delay time of all flights and maximize overall runway throughput. The proposed model incorporates both the technical and regulatory constraints for safety, and system uncertainties. The capabilities of the model are demonstrated in San Diego and Ft. Lauderdale airport runways under various scenarios of different priority logics for queueing, and runway layouts. 2 - Measuring Fuel and Travel Time Benefits for Oceanic Flights through Computer Simulation Arman Izadi, Virginia Institute of Technology, Blacksburg, VA, 24061, United States, armaniz@vt.edu, Antonio Trani Advanced technologies such as Automatic Dependent Surveillance-Broadcast (ADS-B) have significantly improved situational awareness and communication between pilots and controllers in remote oceanic areas. Reducing oceanic separation minima and proposing novel concepts for the Organized Track System could improve oceanic operations. This talk presents a discrete-time model that simulates all phases of flight from takeoff to landing in order to evaluate these concepts in terms of fuel consumption, travel time, controller workload and safety. 3 - OTP Maximizer: Robust Scheduling with Optimal Block and Ground Times Zeynep Sargut, Optym, 7600 NW. 5th Place, Sakarya Caddesi No:156, Gainesville, FL, 32607, United States, zeynepsargut@gmail.com, Pranav Gupta, Revindra K. Ahuja, Nomesh Bolia An airline company usually solves its schedule planning problem considering a little or no stochasticity to maximize its profitability. Highly optimized schedules show an opposite trend in terms of operational metrics such as On Time SC57

88