Understanding Fairness in Unmanned
The future of the commercial drone industry - and the advancement of many new humanitarian and enterprise applications for drone technology - rely upon the safe integration of drones into the airspace. As industry begins to dominate the conversation on drone delivery and other applications, is UTM being developed to ensure that the skies are open to everyone? In this memo, we take a deep dive into fairness in UTM systems.
Open Skies For All
Deconfliction is one of the most practical aspects of a UTM system where fairness is critical. When multiple aircraft have desired routes that are in a conflict with each other, which one gets permission to fly first, and which aircraft will have to make way?
In manned aircraft, “first-come, first-served" is a widely accepted rule for prioritization. For unmanned systems, however, a “first-requested, first-served" approach may have drawbacks in terms of fairness, because of the diversity in types of operations that may want access to the airspace. Some commercial operations like property inspections have the flexibility to plan well in advance; while other applications, like on-demand drone delivery or air taxi services, may not have this flexibility. A first-requested, first-served approach gives unfair priority to those operators that are able to plan ahead. A different approach to prioritization is required to ensure fairness, and there are many factors to consider. The effect of a delay on the outcome of the operation must be taken into account; an inspection may be able to wait a few minutes, when a medical delivery cannot. Access should also be granted in such a way that large players aren’t able to monopolize constrained resources, or simply reserve more airspace than they need to limit competitors' access. Mechanisms must also be included to prevent untruthful behavior from operators seeking to game the system.
Many other complex areas of society have required the development of systems to ensure equity: such as the allocation of radio spectrum for wireless networks, the legislative seat assignments in the U.S. Congress, or the prioritization of patients for organ transplants. These systems may provide some lessons for the development of fairness in UTM; but they don't provide a working model that can be readily ported to the UTM domain. Having evaluated models from structured, points-based schema to market-based mechanisms, Airbus researchers concluded that the nature of a dense airspace is inherently unique: technical, legal, safety, and political aspects of airspace usage make fair resource allocation a challenging problem.
There are, however, metrics that can be used to evaluate a system for fairness. The distribution of delays and number of impacted operations across operators, representing accounting for unmanned operations similar to that performed by manned aircraft, can be evaluated to quantify fairness using formal metrics from the academic literature. These then allow fairness to be balanced against other requirements, such as efficiency, predictability, flexibility, and safety. A variety of different allocation methods could be tried and measured against these criteria, until a solution is identified that meets the needs of all stakeholders.
The airspace isn't yet filled with unmanned vehicles delivering consumer goods to every doorstep. That's just the reason, however, for gathering data now. Collecting as much data as possible on delays and route adjustments in areas where multiple operators are flying now will allow researchers to assess the effect of UTM development decisions on fairness, and ensure equitable and open access to the skies when competition for resources increases.
According to international standards, fairness in UTM is all about access: ensuring that access to the airspace remains fair and equitable. As long as the skies remain relatively empty, access to the airspace isn't a problem. But even if conflicts for resources are rare, they should be handled fairly. And as commercial drone operations scale up and urban air mobility applications become a reality, this need will only increase.
Fairness is something that needs to be built into the system from the start, for important reasons. Unlike regular air traffic management systems, which are managed by a central authority, the decentralized UTM architectures proposed in some regions mean that many individual participants and stakeholders must negotiate access. If participants feel that the UTM framework is fair, they will cooperate; stakeholders who feel that the system is unfair are more likely to act to improve their own situation at the expense of others. Cooperation leads to more efficient use of resources.