Proceedings of the 2017 Winter Simulation Conference W. K. V. Chan, A. D’Ambrogio, G. Zacharewicz, N. Mustafee, G. Wainer, and E. Page, eds. TIME-PARALLEL SIMULATION OF AIR TRAFFIC NETWORKS Young Jin Kim Artificial Intelligence Products Group Intel Corporation Santa Clara, CA 95054, USA Dimitri Mavris School of Aerospace Engineering Georgia Institute of Technology Atlanta, GA 30332, USA Richard Fujimoto School of Computational Science and Engineering Georgia Institute of Technology Atlanta, GA 30332, USA ABSTRACT Air traffic management is widely studied in several different fields because of its complexity and criticality to a variety of stakeholders. However, the exploding amount of air traffic in recent years has created new challenges to ensure effective management of the airspace. A fast time simulation capability is essential to effectively explore the consequences of decisions of air traffic management. A new algorithm for simulating air traffic networks using a time-parallel simulation approach is proposed that distributes time segments of the simulation scenarios across different processors. A simulation model for the National Airspace System (NAS) is described and validated. The components of the simulator are described as well as the parallel simulation algorithms. Experimental results utilizing real-world traffic data for the continental U.S. are presented demonstrating the speed ups achieved by a prototype simulator. These results illustrate that time-parallel simulation can be used to significantly accelerate certain air traffic simulations. 1 INTRODUCTION According to the Federal Aviation Administration (FAA)’s latest forecast, domestic enplanement will be increased by 1.45 times over the next twenty years. Over the same time period, it is estimated that the number of passengers taking international flights into or leaving the U.S. will be double (Aerospace Forecast, FAA 2016). This increased traffic could result in significant delays in the National Airspace System (NAS). According to one study (Ball et al. 2010), air traveler delays accounted for approximately $33 billion in direct or indirect costs to passengers, airlines and other parts of the NAS in 2007. Computer simulations of the NAS play a key role in helping to alleviate this concern by evaluating many different possible scenarios and situations. Rapid execution of simulation models is important in order to explore a wide variety of scenarios quickly. Parallel processing offers an approach to accelerate simulation executions, and several different parallel simulation algorithms have been explored. These approaches use spatial parallelism where one divides the NAS into distinct regions, and one distributes the state and associated computations to transform this state across different processors so that they can be performed concurrently. To ensure that the computation is correct, it is necessary to properly synchronize these computations. In an air traffic network simulation, the components of the system such as airports, air traffic control centers, flights etc. for a region are typically mapped to a single logical process (LP). Each LP computes its internal states during the simulation run and communicates necessary events with other 1013 978-1-5386-3428-8/17/$31.00 ©2017 IEEE