1

Introduction

The FAA Extension, Safety, and Security Act of 2016 called for the Federal Aviation Administration (FAA) to “enter into an arrangement with the National Academies to study the potential use of probabilistic assessments of risks by the Administration to streamline the integration of unmanned aircraft systems into the national airspace system, including any research and development necessary.” The FAA and the National Academies of Sciences, Engineering, and Medicine agreed that because various types of risk assessments were being used, depending on types of vehicles and methods of operation, the National Academies could best assist the FAA by looking at a broader set of risk issues. The Committee on Assessing the Risks of Unmanned Aircraft Systems (UAS) Integration was formed to evaluate the potential of probabilistic assessments of risks and other risk assessment methods for streamlining the process of safely integrating unmanned aircraft systems into the National Airspace System and identifying supporting research and development opportunities in this field.

In undertaking this study, the committee considered recent, current, and planned FAA efforts to evaluate the risks associated with the integration of UAS into the National Airspace System and risk assessment methods. It also considered mechanisms for assessing severity and likelihood metrics required for probabilistic and other appropriate risk assessment methods based on UAS design characteristics (e.g., weight, speed, materials, and technologies) and operational characteristics (e.g., airspace characteristics, population density, and whether UAS are piloted remotely or autonomously). The committee also sought to determine how the scope and detail required of risk assessment methods may vary for different sizes and operations of UAS (e.g., Part 107 versus Part 91 operations) or whether a certain class of UAS (micro, etc.) could be approved to operate with the assumption they are inherently low risk. In addition, the committee sought to evaluate other methods that could reasonably be used to evaluate the risks of UAS integration in the National Airspace System.

The committee was guided by a number of questions, such as the following:

• What are the benefits and limitations of these alternative risk assessment methods? How do these alternative methods compare to probabilistic risk analysis methods as well as severity and probability metrics traditionally used by the FAA for manned aircraft?
• What state-of-the-art assessment methods are currently in use by industry, academia, other agencies of the U.S. government, or other international civil aviation authorities that could benefit the FAA?
• What are the key advancements or goals for performance-based expanded UAS operations in the National Airspace System that can reasonably be achieved through the application of the recommended risk assessment methods in the short term (1-5 years), mid-term (5-10 years), and longer term (10-20 years)?

• What are the key challenges or barriers that must be overcome to implement the recommended risk assessment methods in order to attain these key goals?

In light of ongoing research and likely advances in risk assessment methods by other organizations, the committee also considered what research and development projects related to risk assessment methods should be the highest priority for the FAA. Last, the committee investigated whether there are other related recommendations to streamline FAA processes (not governed by regulation) that would either improve the effectiveness of risk assessment methods for integration of UAS into the National Airspace System or expedite the development of such methods. (The committee’s full statement of task is included as Appendix A.)

The committee was able to consider and comment on the effectiveness of risk assessment methods as they pertain to decision making and different modes of UAS operations. However, the committee does not recommend changes to regulations governing UAS operations, nor does the committee recommend changes to the organization of the FAA. The scope of this study includes UAS certification as well as operational approval.

The committee has concluded that an evolution of current FAA risk assessment methodologies is needed to integrate UAS into the National Airspace System in a timely yet safe manner. A principal driver of this conclusion is the wide variety and number of UAS operations in tandem with societal safety-related benefits that those operations can provide the public.

UAS operations vary from (1) those under Part 107 (i.e., the UAS weighs less than 55 pounds, it is within visual line of sight of an operator who is operating only that UAS, it is operating at a maximum altitude above ground of 400 feet, and it is not over people not participating in the operation); to (2) low-altitude micro UAS and small UAS operating beyond visual line of sight in rural areas; to (3) UAS operating beyond visual line of sight at low altitude over people at varying population densities, including in cities; to (4) large UAS whose missions take the platforms into controlled airspace at en route altitudes.

Applying probabilistic risk analysis methods developed over several decades for operations of manned aircraft, from which huge amounts of operational data are available, to the full range of UAS operations does not take into account either fundamental differences from manned aviation present in most UAS operations, particularly low-altitude operations, or the relative youth (compared to manned aviation) of the UAS industry and lack of operational data.

Figure 1.1 is illustrative of the aforementioned societal safety-related benefits, when an emergency flotation device was dropped to swimmers in danger. There are numerous additional examples: support for emergency responders, such as safer disaster assessment; improved safety and effectiveness in infrastructure assessment, such as for cellular telephone towers and railroad rights-of-way; emergency delivery of medicine; and reduction in highway accidents and environmental pollution resulting from safe UAS delivery of packages. Such safety-related societal benefits are additional to, and in many cases independent of, any economic benefits that UAS operations might have.

The U.S. military has extensive experience with UAS, both large and small, but to date that experience has not translated well into the integration of civilian UAS into the National Airspace System. Military UAS usually operate in airspace segregated from manned airspace. Many UAS operations take place in airspace over ground troops, and those operations certainly pose some risk to those soldiers. The military conducts such operations, however, in part because the magnitude of that risk is balanced against the much more substantial risk that soldiers face in combat and the ability of UAS to mitigate that risk and support mission success.

Current FAA probabilistic risk analysis methodologies do not take these societal safety-related benefits into account. Any UAS operation is therefore viewed as increasing the risk only to the National Airspace System, and improved safety for swimmers, firefighters, railroad operators, drivers, or others is irrelevant. Such a viewpoint can make the process of UAS integration into the National Airspace System start off with a negative bias.

One of the major issues facing the FAA is cybersecurity. Cybersecurity is one of many factors to be considered in any probabilistic risk analysis, and although the committee was aware of issues related to cybersecurity, it was not tasked to focus on them. The committee believes that the FAA, industry, and operators are aware of the importance of cybersecurity and are seeking to address this subject in many different ways.

In developing its recommendations, the committee met three times, receiving presentations from the FAA,

academia, and industry related to probabilistic risk analysis for UAS operations. The list of speakers for these meetings can be found in Appendix C.

Chapter 2 provides background information to the committee’s recommendations, including assumptions and guiding principles that the committee formulated, and a set of key definitions that the committee adopted for this report.

Chapter 3 summarizes the committee’s assessment of current FAA practices with regard to probabilistic risk analysis for UAS operations, to include the FAA safety culture and risk assessment processes.

Chapter 4 presents the bulk of the committee’s recommendations for evolving the FAA decision-making paradigm for assessing risk of UAS operations. The chapter addresses developing a more appropriate risk analysis process, how decisions are best driven with data for a young UAS industry, and the delegation of certain levels of risk analysis by the FAA to the private sector.

Chapter 5 discusses the committee’s recommendations with regard to recommended research areas related to probabilistic risk analysis for UAS operations.

The committee made a number of assumptions in developing its findings and recommendations and discussed common definitions and points of reference. They form the basis of Chapter 2.