1

Introduction

Adversaries compete with each other strategically and tactically. Historically, the tactical edge of the battle space has been where two armies meet face to face or two navies engage on the high seas. As technology has advanced, the definition of the tactical edge of the fighting forces has evolved as well. When a remotely piloted vehicle over a contested region can be controlled from thousands of miles away while small special forces units deep in enemy territory can harness the power of space to leverage their capabilities, the idea of a crisply defined tactical edge becomes difficult. Nonetheless, it still exists, and there are aspects of the tactical edge that are significantly different from what would be normally found in a garrison or at a fixed base. These challenges include providing power to the increasing number of information technology (IT) solutions embedded in deployed units and systems.

The vision of the near-future tactical battle space is one populated with IT-enabled devices:

• Swarms of unmanned systems in the air, on the land, and under the water;
• Warfighters augmented with smart everything: situational awareness systems, physiological monitoring systems, real-time tracking and mapping, and more;
• Sensors deployed everywhere, some camouflaged as terrain features, some embedded in operational vehicles, and some more conventionally static;
• Artificial intelligence (AI) and machine learning (ML) applications enabling speedy decision making;

• Strategic assets, such as satellites and high-altitude loitering aircraft (some manned, some not), monitoring activity and providing data to central control facilities; and
• Military Internet of things (MIoT) devices acting in a web of smart collaboration.

Each of these capabilities brings a staggering amount of data and intelligence to the activity of conflict, enabling a level of precision and collaboration that would have been unimaginable 50 years ago and that now is somewhat difficult to truly comprehend in terms of capabilities and implications.

The U.S. Air Force (USAF) uses a lot of energy. It accounts for 48 percent of the total Department of Defense (DoD) energy expenditures. Within the USAF, 86 percent of the expenditures are associated with aviation fuel.¹ Energy usage by installations accounts for only 11 percent of the energy usage. In the context of these very large numbers, the energy needed for tactical-edge computing is tiny by comparison. And yet, the data generated, used, and communicated in and from the tactical edge is becoming a critical element of the future operating environment.

The ongoing revolution in information technologies, particularly in small appliance-type devices, has enabled this proliferation of capabilities that contribute to efficiency and effectiveness of information-rich activities. Each of these devices, however, requires energy in order to be functional. On an individual device level, the energy needs are typically very small. Further, energy efficiencies continue to improve along several architectural pathways: reduced energy usage reduces operational heat, battery replacement frequency, and electromagnetic radiation. But as the utility of these small devices continues to be proven, more and more of them proliferate in operational environments, leading to aggregate energy usage that can be significant.

The new great power competition² will require that the USAF, like the other services, adopt product and process technologies that ensure it is able to compete with adversaries in very fast decision cycles. This need requires active information collection and processing to support decision making, collaborative operations, and multidomain operations. The new great power competition includes rivals that are peer-level or near-peer-level technological forces as well as nonstate actors associated with the ongoing “war on terror.” It is also characterized by the increasing use of robots, unmanned vehicles of all types (in the air, at sea, in space, and on land), and the insinuation of artificially intelligent computational agents into

___________________

¹ See Air Force Civil Engineer Center, 2017, “Energy Flight Plan 2017–2036,” https://www.safie.hq.af.mil/Portals/78/AFEnergyFlightPlan2017.pdf?ver=2017-01-13-133958-503.

² See T. Lynch III (ed.), 2020, Strategic Assessment 2020: Into a New Era ofGreat Power Competition, NDU Press: Washington, DC, https://ndupress.ndu.edu/Publications/Books/Strategic-Assessments-2020/.

both administrative and weapon systems. Thus, the information dimension of competition continues to increase in both importance and power, requiring the proliferation of information infrastructure elements to all ends of the force projection continuum.

The vision is compelling, and the planning is impressive. The energy needed to fuel the vision, however, seems to have taken a back seat in the planning process. For the fixed installation information infrastructure planning, this may make sense, because the energy needs of a fixed installation are well understood and managed. The planning for future operations does include dynamic basing as well as the more conventional tactical-edge components, for which energy planning becomes more critical. One question that requires addressing as the planning proceeds is: how much energy will be needed to power the tactical-edge information infrastructure and will that energy be readily available?

Concern over these matters prompted the Air Force to ask the National Academies to conduct a study to examine how energy needs for the information processing needs at the tactical edge are being considered and, if appropriate, to recommend ways in which such planning could be improved.

STUDY PROBLEM

Will the Air Force be adequately positioned to meet the energy needs associated with data-driven operations at the “edge” in the year 2030? To answer this overarching question, it is necessary to parse it into component issues.

• First, what will the energy needs be at the tactical edge to support data-driven operations? This question itself spawns many further questions, such as what the expected technology base will be; how energy-efficient computational approaches will be implemented, if at all; and what the energy needs of the edge infrastructure will be, including environmental controls for both humans and machines and the local data transfer between small devices. All of these considerations are dependent on the amount and types of data expected to drive the vision of the future battlefield.
• Second, how will energy provision at the edge be prioritized among users in the face of competition for scarce resources or in the face of attack on energy supplies? This question spawns further questions as well, including what is known about adversarial plans or capabilities to attack energy systems; how resiliency is being engineered into the energy at the edge infrastructure; and planning expectations for edge data-driven operations under energy scarcity.
• Third, does the USAF currently have or plan to have adequate human resources to build, deploy, and maintain the energy infrastructure for data-driven opera

• tions at the edge? What are the expertise requirements for technical support to data-driven operations at the edge?

These three component questions served as the guiding thoughts in developing a study plan that addressed each element of the statement of task.

STATEMENT OF TASK

The Air Force Studies Board at the National Academies of Sciences, Engineering, and Medicine will establish an ad hoc committee to investigate energy challenges and opportunities for future data-driven operations in the USAF.

The committee will plan and convene a multiday workshop that would investigate and discuss the energy challenges and opportunities of the USAF’s future operating environment. Workshop participants shall examine what steps and plans the USAF is taking and/or should be considering now in order to successfully develop, deploy, and sustain the weapon systems needed to compete in an emerging information-rich environment. The workshop shall also investigate energy demands for weapons systems from the airbase to the battle space, including information requirements and needs associated with capture, curation, storage, exploitation, and transmission of energy to enable the deployment and operation of data-reliant systems. The committee shall develop the workshop agenda, invite speakers and other attendees, moderate discussions, and provide resulting meeting materials.

The committee will plan and conduct an in-depth study that would:

1. Investigate the current state of Air Force planning, research and development, and expectations related to energy usage for military operations in the 2030 time frame.
2. Investigate potential threats to energy assurance and access based on recent events and assumptions of future energy dependencies that should inform Air Force/government planning for energy generation, storage, and use.
3. Investigate and describe current research and state of the art in energy-efficient computation, including hardware, software, and big data.
4. Investigate and describe the energy needs for advanced weapons platforms, including static infrastructure that provides support to machine learning, artificial intelligence, and integrated operations.
5. Recommend manpower, research, and expertise requirements needed for the future energy environment.

The committee shall convene meetings of the study committee and other attendees, as required, to gain relevant information. The committee shall also provide a report summarizing the results from this study.

REPORT AUDIENCE

This effort was requested by the Deputy Assistant Secretary of the Air Force for Science, Technology, and Engineering (SAF/AQR), and its contents are intended to advise senior leadership responsible for weapons systems sustainment, equipment, and logistics and installations resource requirements, including research. Because the challenge of understanding and solving energy needs for IT at the tactical edge crosses many different specialties, ranging from communicators to civil engineers, this study addresses a broad range of issues. As such, different parts of this report are expected to be of more interest to some offices than to others. However, it is a fundamental conclusion of this report that coordination of efforts across disciplines and responsibilities is a critical element to long-term success in meeting the needs of the future.

THE COMMITTEE’S APPROACH

This effort was initiated in November 2019, and the committee first convened in February 2020. Members of the interdisciplinary committee served as volunteers for the duration of the effort. The committee members were selected and appointed by the National Academies, which focused on ensuring that there was a breadth of expertise, including both researchers and practitioners. Committee members brought expertise in computer science, computer engineering, information security, intelligence, alternative energy sources, military operations, human-systems integration, and engineering economics. See Appendix E for committee member biographies.

To conduct its study and develop consensus for the final report, the committee intended to travel to and meet with a broad range of organizations and individuals. Because of the SARS-CoV-2 (COVID-19) pandemic restrictions, which went into effect in March 2020, the committee held all meetings with subject matter experts virtually. The inquiry began with a 3-day workshop held in April 2020. The proceedings of this workshop are included in Appendix C of this report. Subsequent to the workshop, Zoom³ meetings were held with a variety of authorities and stakeholders over a period of 6 months. The agendas for the committee’s

___________________

³ Zoom is a publicly traded corporation, headquartered in San Jose, California, that provides software to enable virtual conferencing capabilities. See Zoom, “About,” https://explore.zoom.us/about.

data-gathering sessions are listed in Appendix B. Presentation materials used are available, by request, from the National Academies through the public access file.

The committee thanks all presenters for the frank and open discussions, particularly in light of the difficult circumstances. The committee is particularly grateful to those who were suffering from COVID-19 and still consented to participate with the study inquiry. Their commitment to progress and analysis is commendable. The committee felt that it was important to verify that the subject of study was in fact an important topic and, as a result, made it a practice to ask each speaker some version of the following question:

The answers received back were universally affirmative on both counts: the changes occurring in the uses of data processing at every level are causing unforeseen challenges that need to be considered. The committee came away from these meetings with a very deep appreciation for the thoughtfulness, inventiveness, and dedication of those working in and for the U.S. government to address these challenges. The future is in good hands.

REPORT ORGANIZATION

This report is organized in a way that mirrors the task given, with the material presented in each section to highlight the findings and conclusions developed by the committee. This chapter has presented the rationale and methodology for the study. The remaining portions of the report are as follows:

• Chapter 2 provides an analysis of the problem, a discussion regarding the potential threats to energy provision, and energy dependencies and demand issues. The current state of the art in energy-efficient computation and architecture is presented, as is a discussion on energy needs from an enterprise perspective.
• Chapter 3 summarizes the findings and recommendations derived from the data collection and analysis and suggests a priority order for addressing the recommendations.
• The report concludes with several appendixes that provide additional detail on several aspects of the report, including the study’s data gathering activities and committee member biographies.