Bruce J. Holmes

Air Transportation Strategic Trade Space Modeling and Assessment Through Analysis of On-Demand Air Mobility with Electric Aircraft

1 Manager of Strategic Analysis, Aeronautics Research Mission Directorate, NASA HQ, Washington DC 20546 2 Chief Executive Office, 205 Skimino Landing Dr., Williamsburg, VA 23188, AIAA Fellow 3 Chief Science Officer, 508 Olive St., Santa Cruz, CA 95076, AIAA Member 4 Chief Technology Officer, 784 Rosewood Dr., Palo Alto, CA 94303, AIAA Member 5 Senior Economist, GRA, Inc., 115 West Av, Suite 201, Jenkintown, PA 19046 6 Senior Consultant, LMI, 2000 Corporate Ridge, McLean, VA 22102, AIAA Senior Member 7 Senior Fellow, LMI, 2000 Corporate Ridge, McLean, VA 22102 8 Senior Fellow, LMI, 2000 Corporate Ridge, McLean, VA 22102 9 Director of Transportation Research, Old Dominion University, 5115 Hampton Blvd, Norfolk, VA 23529, AIAA Fellow 10 Metron Aviation, 45300 Catalina Court, Suite 101, Dulles, VA 20166 11 Metron Aviation, 45300 Catalina Court, Suite 101, Dulles, VA 20166 12 IAI, Inc., 15400 Calhoun Drive, Suite 400, Rockville MD 20855 13 NASA Langley Research Center, Hampton, VA 23681 14 NASA Langley Research Center, Hampton, VA 23681, AIAA Senior Member 15 Associate Technical Fellow, Boeing Commercial Airplanes, Seattle, WA 12th AIAA Aviation Technology, Integration, and Operations (ATIO) Conference and 14th AIAA/ISSM 17 19 September 2012, Indianapolis, Indiana AIAA 2012-5594

Modal Preference Modeling of Transportation Demand and Supply for Strategy Portfolio Analyses - Results and Future Plans

Future demand for transportation is and will continue to be shaped by forces that have not been well accommodated in past strategic analyses; further, regression-based analytical methods are less well suited than generative methods for projecting demands for modal options that have little historical data on which to base regressions. New transportation modes, business models, consumer behaviors and vehicle capabilities are the primary factors not well managed in regressive demand projection methods. An example is the ability to study co-evolutionary effects such as a “virtual” mode or modes (i.e., interacting by telepresence as a modal option) on population dynamics or urbanization trends. The risk is that current national transportation strategies in air mobility tend to be constrained by “business as usual” considerations of a rear-view-facing nature. In addition, air transportation demand projections are frequently made in modal isolation; that is, projections for air travel demand have not typically accounted for the full context of all other existing and prospective new modal options and their improvements. Further, if the strategy development processes do not consider the prospect of vastly different characteristics of external context, including new consumer behaviors and modal options then the strategies carry inherent risks. The plausible ranges and combinations in trends or vectors in technologies, energy, environmental, and prosperity considerations comprise a wide range of future conditions s in which strategies must be evaluated. Because 1 Chief Executive Office, 205 Skimino Landing Dr., Williamsburg, VA 23188, AIAA Fellow 2 Chief Science Officer, 508 Olive St., Santa Cruz, CA 95076, AIAA Member 3 Chief Technology Officer, 784 Rosewood Dr., Palo Alto, CA 94303, AIAA Member 4 Director, Transportation Research, Old Dominion University, 5115 Hampton Blvd, Norfolk, VA 23529, AIAA 2 Chief Science Officer, 508 Olive St., Santa Cruz, CA 95076, AIAA Member 3 Chief Technology Officer, 784 Rosewood Dr., Palo Alto, CA 94303, AIAA Member 4 Director, Transportation Research, Old Dominion University, 5115 Hampton Blvd, Norfolk, VA 23529, AIAA Fellow 5 Senior Consultant, LMI, 2000 Corporate Ridge, McLean, VA 22102, AIAA Senior Member 6 Senior Fellow, LMI, 2000 Corporate Ridge, McLean, VA 22102 7 Senior Economist, GRA, Inc., 115 West Av, Suite 201, Jenkintown, PA 19046 8 Manager of Strategic Analysis, Aeronautics Research Mission Directorate, NASA HQ, Washington DC 20546

Complexity Science Tools for Interacting 4D Trajectories and Airspace Phase Transitions

ly, the airspace can be thought of as containing a gas of dynamical trajectories, continuously interacting and replanning. At low density the interaction is minimal, a kind of ideal gas where all the trajectories are easily separated from each other. At higher density, this gas may go through a phase transition(s), losing degrees of freedom, and condensing into a more regular and rigid form, unable to find ways to keep its components safely separated from each other without large inefficiencies. This approach improved the understanding of overall properties of the airspace. For example, we were able to characterize capacity limits and forecast congestion and other suboptimal behaviors before they occur, etc. With this ontology, we examined the phase structure of the airspace using tools drawn from the science of traffic physics, in search of possible phase transition structure, as well as precursors to these phase transitions. A. 5DT Dynamical Trajectories At the most elementary physical level, the airspace consists of air, aircraft and obstacles (weather cells, closed airspace, etc.). However, since aircraft move over time, it was logical to up-level these dynamical moving aircraft to the abstraction of trajectories. This trajectory abstraction proved very useful in representing many issues in airspace design and management. In our analysis, we went up one step higher still: our trajectories themselves were considered dynamical entities. Our trajectories themselves changed over time. Trajectories were continuously (in practice, every small discrete Δt) re-calculated (replanned) while the aircraft was in flight. This replanning is made essential by the combination of an interacting system of trajectories combined with an evolving system of information. (Constraints, such as weather or unforeseen flight alterations, can emerge over time). A subtle creation of this up-leveled abstraction of dynamical trajectories and adaptive replanning was two dimensions of time. There was the time endemic to the passage of origin-to-destination time within trajectories, namely flight time. There was also an additional Meta Time over which the trajectories themselves changed too. This approach could also be seen as “from” time and “to” time. The state of the airspace at a given future time may change depending on what time was used to compute and forecast from, as new information is constantly arriving. Thus, with three geometric parameters, x, y, and z, plus two dimensions of time, the term 5DT represents an appropriate characterization of our approach to trajectory management. Figure 1-1 shows an artist’s rendition of a 5DT trajectory. Over time, the trajectory itself is deformed according to physics-like “forces” exerting pressure on the trajectory, thus changing its shape. The deformation might be to achieve minimum separation or to avoid weather. T

Complexity Science Applications to Dynamic Trajectory Management: Research Strategies

The promise of the Next Generation Air Transportation System (NextGen) is strongly tied to the concept of trajectory-based operations in the national airspace system. Existing efforts to develop trajectory management concepts are largely focused on individual trajectories, optimized independently, then de-conflicted among each other, and individually re-optimized, as possible. The benefits in capacity, fuel, and time are valuable, though perhaps could be greater through alternative strategies. The concept of agent-based trajectories offers a strategy for automation of simultaneous multiple trajectory management. The anticipated result of the strategy would be dynamic management of multiple trajectories with interacting and interdependent outcomes that satisfy multiple, conflicting constraints. These constraints would include the business case for operators, the capacity case for the Air Navigation Service Provider (ANSP), and the environmental case for noise and emissions. The benefits in capacity, fuel, and time might be improved over those possible under individual trajectory management approaches. The proposed approach relies on computational agent-based modeling (ABM), combinatorial mathematics, as well as application of “traffic physics” concepts to the challenge, and modeling and simulation capabilities. The proposed strategy could support transforming air traffic control from managing individual aircraft behaviors to managing systemic behavior of air traffic in the NAS. A system built on the approach could provide the ability to know when regions of airspace approach being “full,” that is, having non-viable local solution space for optimizing trajectories in advance.