1 Analyzing Air Taxi Operations from a System-of-Systems perspective using Agent-based Modeling Julian R. Archer 1 , Alden W. Black 2 , and Satadru Roy 3 Purdue University, West Lafayette, IN, 47907 The purpose of this paper is to introduce air taxi operations as a system-of-systems type problem. Five primary traits, operational and managerial independence, geographical distribution, emergent behavior, networks, and trans-domain are highlighted since they call attention to a network-of-systems dynamic behavior focus as opposed to an individual system, static behavior focus. More specifically, this paper tries to extend work done by Mane and Crossley 1 in an attempt to capture a more realistic air taxi operation via agent- based modeling using Matlab. Our interest is to understand which particular mixture of aircraft types, as selected for our preliminary analysis, are most viable for profitable air taxi operations between random city pairs given emergent business traveler demands. Mane and Crossley 1 identified the Eclipse 500 and Cirrus SR-22 as two of the aircraft most promising for air taxi operations. However, our results revealed the Piper Meridian, Daher-Sacota TBM 850, and Pilatus PC-12/47 single engine turbo-props as the most efficacious aircraft. I. Introduction he notion of air taxi operations isn’t new. In fact, the idea has been studied by numerous researchers at various research institutions, for instance: 1, 2, 3, 4, 5, 6, 7, and 8 . The goal of air taxi operations in simple terms is to utilize small aircraft (say single piston engine, twin piston engine, turboprop, or very light jets) to provide an on-demand “point-to-point” air travel service. The “point-to-point” service we referred to would be from a location (point) near the travelers’ origin to a location near their desired destination. The common agenda in understanding the dynamics that are involved in air taxi operations is the fact that it can serve as an alternative mode of transportation to the traditional commercial airline service, which requires passengers to: travel to a major airport via ground transportation to eventually board an aircraft; fly from that major airport to a hub where a connecting flight can be made; fly from the hub to a major airport within a reasonable radius of the passenger’s destination; utilized ground transportation once again to reach the desired destination. Air taxi operations are ideally an alternative that would reduce the travel time involved in moderate-to-long-distance ground (door-to-door) travel and reduce the hassle of using commercial airline services, for instance seemingly-inefficiently travel from the south of the US to the Midwest to “connect” before finally travelling to the northeast. The air taxi idea at first seemed relatively straight forward. Unfortunately it is not, and there has been much debate on the marketability and viability of its operations, as cited by Mane and Crossley 1 : “On one side of the debate, several experienced airline executives and even a former director of an aircraft manufacturer have proposed and / or launched air taxi services. On the other side of the debate, aviation business analysts claim that air taxi service would not reach a sufficient share of the market to be successful”. To understand the debate we should realize various factors play a role in the determining whether the air taxi operation turns out to be a marketable and viable one, or fails. For instance, the service is not able to fly everywhere. Therefore, which airports and routes should be served? According to Boyd et al. 3 , the airports to be served would need to be nearby popular destinations and be easily accessible to ensure adequate passenger demand; also, the routes to be served between city pairs should not exceed 350 miles in point-to-point distance. Other questions of concern would be “What are acceptable passenger wait times if they show up?” and would like a flight immediately, and “Should passengers be flown within a certain window period (e.g. within 1-2 hours of arriving at the airport) or immediately on arrival?” More interestingly is the issue of aircraft type. What types of aircraft and how many should we use and dispatch? We should realize that the dispatching method directly affects the number of deadhead flights during operations. 1 Graduate Student, School of Industrial Engineering, Email archerj@purdue.edu, and AIAA non-member 2 Graduate Student, School of Aeronautics and Astronautics, Email black36@purdue.edu, and AIAA non-member 3 Graduate Student, School of Aeronautics and Astronautics, Email roy10@purdue.edu, and AIAA non-member T