Addressing and Amending “Attritability” of Unmanned Vehicular Systems: Why Cost as a Strategy Undermines National Power

Abstract

This essay posits that the current definition of attritability as relevant to the viability and value of employing unmanned vehicular systems (UVS) is laden with inaccuracies that can affect calculation(s), planning, tolerances and thresholds for military operational engagements.  To this point, a revised definition is proposed that affords clarified insight to the multiple variables that should be regarded in any genuine considerations of UVS deployment and relative operational, tactical and strategic worth. 

Introduction

The emergence of unmanned and increasingly autonomous vehicular and weapon systems has altered both the tactical conduct of military operations, and the strategic calculus by which force is considered, authorized, employed, and sustained. In prior work, we have posited that these systems possess the capacity to alter the psychological, political, operational, and ethical thresholds for the initiation, continuation, and escalation of conflict. Their ability to reduce immediate risks to friendly personnel, accelerate operational tempo, and distribute combat capabilities across multiple domains changes the ways that military and political leaders perceive costs and benefits of employing force. Yet, as these systems continue to proliferate, another concept, attritability, has emerged that warrants considerably deeper examination.

Although attritability has become a commonly invoked descriptor within contemporary defense planning, acquisition, and force design, it is frequently employed as a shorthand for relatively inexpensive platforms that can be accepted as expendable during military operations. Such a definition is incomplete, if not inapt. We argue that limiting attritability to the monetary replacement cost of a platform risks misunderstanding its broader strategic implications and, in so doing, may inadvertently lower thresholds for conflict in ways that are neither anticipated nor adequately governed.

Addressing Attritability 

 A more realistic and operationally useful conception of attritability must recognize that the construct extends well beyond the physical loss of an individual platform, and thus should be understood as a multidimensional assessment of the military, political, economic, informational, industrial, and strategic consequences associated with both accepting the degradation or loss of an unmanned capability, and the effects that such loss incur in other domains and dimensions of military force.

 Such an expanded definition provides a more accurate representation of the actual costs incurred by employing ostensibly “expendable” systems. From an operational perspective, the destruction of an unmanned system is limited to the hardware through which it operates. Modern semi-autonomous and autonomous platforms entail sophisticated sensing architectures, communications networks, software, artificial intelligence (AI) systems, mission data, and electronic warfare capabilities. Their destruction may reveal sensitive technologies to adversaries, expose operational concepts, enable reverse engineering, and/or compromise intelligence collection methods. Consequently, the true cost of attrition must be regarded to include losses of information superiority and technological advantage that may substantially exceed the economic replacement value of the platform itself.

Similarly, industrial considerations are central to any comprehensive definition of attritability. A platform may be relatively inexpensive to manufacture, but may depend upon fragile supply chains, scarce materials, specialized semiconductors, or limited production volume. Thus, the capacity to replace lost systems at operationally relevant rates becomes as important as their unit acquisition cost. In this light, attritability must incorporate considerations of industrial resilience, manufacturing scalability, logistics, and sustainment as integral components of the construct.

Political dimensions likewise deserve careful consideration. As we have heretofore noted, a frequently cited advantage of unmanned systems is their capacity to reduce friendly casualties, thereby diminishing domestic social and political sensitivity to military operations. While this characteristic may increase strategic flexibility, it may also lower perceived barriers to employing force. If such losses are perceived as politically tolerable (i.e.- because human casualties are minimized), there exists a corresponding risk that military options become increasingly normalized as instruments of statecraft. In this regard, attritability intersects directly with our previous observations regarding evolving thresholds and tolerances for war. The more disposable military capabilities appear, the more readily they may be committed to operations whose long-term strategic consequences remain inadequately appreciated.

Thus, there are important cognitive and organizational dimensions that must be leveraged. Operators, commanders, and political (and social) leaders may become relatively desensitized to platform losses when such losses are anticipated by design. This normalization of attrition can alter institutional risk calculations, influence campaign planning, and encourage operational approaches emphasizing quantity over strategic discrimination. Such cognitive shifts may subtly reshape military culture and strategic decision-making absent a corresponding consideration of cumulative escalation risks.

Importantly, attritability also must be understood relationally (versus absolutely). A platform may be highly attritable in one operational environment yet strategically indispensable in another. Likewise, acceptable rates of attrition can and often do vary according to mission objectives, adversarial capabilities, alliance commitments, public perceptions, and broader geopolitical circumstances. Consequently, attritability should be regarded as a contextual assessment emerging from the interaction of technological characteristics, economic costs, operational requirements, strategic objectives, and political realities. To be sure, the Department of War (DoW) has embraced the term “attritable” with enthusiasm that borders on sloganry. Despite repeated congressional direction (e.g.- most recently reflected in 2025 and  2026 National Defense Authorization Act (NDAA) committee reports) – that the term remains undefined and ambiguous, it continues to serve as a central organizing concept for major unmanned and autonomous initiatives. Years into the Replicator initiative and informed by Ukraine’s drone-swarm experience, the word remains yet undefined in statute, regulation, or formal DoW policy. To this point, the House Armed Services Committee  has noted that the Department’s use of “attritable autonomous” terminology “remains ambiguous,” carrying forward concerns from prior cycles. However, funding continues to flow while requirements writers, program managers, and industry designers operate under mismatched expectations. This ambiguity risks transforming a potentially powerful enabler of mass force and resilience into strategic self-deception at a time when the United States (U.S.) must prepare for protracted, high-consumption conflict against near-peer adversaries capable of imposing severe economic and industrial costs across all instruments of national power.

Amending Attritability: A More Refined Definition

To address this issue, herein we propose a refined definition of attritable systems; viz.- physical platforms and their integrated mission packages that are purpose-designed to accomplish assigned missions and return, when possible, yet whose loss is both operationally and economically acceptable because they can be rapidly produced and replenished at combat tempo through resilient industrial surge. We offer that what makes this definition unique is its explicit grounding in replenishment realism and a multi-factor spectrum rather than a narrow, single axis focus on unit cost. By directly filling the persistent statutory and policy void repeatedly highlighted by Congress, it provides a rigorous, auditable foundation that has been missing from recent acquisition practice. We opine that to resolve this ambiguity and enable more honest decision-making, the Secretary of War should direct formal adoption of this refined definition across Department of War acquisition guidance, service instructions, and future National Defense Authorization Act reporting requirements. Such adoption would establish a shared, precise language for requirements development, industrial planning, and program oversight, while creating a common standard that unites U.S. partners and allies in harmonizing attrition-tolerance criteria, thereby strengthening coalition interoperability, collective supply-chain resiliency, and integrated deterrence.

We offer that the proposed seven-level attrition-tolerance scale and supporting decision matrix operationalize this definition for any military system or asset (e.g. – munitions, unmanned platforms, surface vessels, capital ships) by anchoring classification in criteria for operational risk tolerance and replenishment realism. By structuring discussion around these principles, the framework advances the 2025 National Security Strategy (NSS) imperatives for technological overmatch, resilient supply chains across international borders, economic security as national security, and integrated deterrence with allies and partners in an era of strategic competition.

Why a Clear(er) Definition Is Critical

For the DoW and Congress, an explicit multi-factor framework can sharpen requirements development, program oversight, and resource allocation. Government Accountability Office (GAO) 24-106831 and GAO-25-107569 assessments have documented how vague or shifting requirements often contribute to schedule delays, cost growth, and technical risk.  Current ambiguity surrounding the term and definition of “attritability” has exacerbated these challenges. A standardized framework grounds programs in realistic budget, production surge, and replenishment timelines instead of issuing vague “low-cost attritable” mandates that can later collide with critical materials chokepoints, fragile sub-tier supply chains, and/or protracted software recertification cycles.

For industry (i.e.- prime contractors and sub-tier suppliers) we believe that a clear 1 – 7 attrition-tolerance scale and decision matrix ( as depicted in Tables 1 and 2, respectively) can mitigate  guesswork by allowing designers to target precise levels of capability, benefit and risk (whether rapid commercial-off-the-shelf expendables or mid-tier systems engineered to tolerate loss) while protecting sensitive payloads and thus minimizing multidimensional loss. This alignment accelerates proposal quality, reduces contracting friction, increases competition (including from non-traditional suppliers), lowers pricing through reduced uncertainty, focuses capital investment on resilient domestic and allied supply chains, and enables the rapid industrial scaling essential for success in protracted near-peer conflict.

Table 1

The proposed framework employs a universal seven-level scale grounded in operational risk tolerance and replenishment realism, as follows:

A 1 – 7 Attrition-Tolerance Scale

Note that Levels 4–6 (Light to Heavy Attritable) entail diffusion-of-responsibility risks inherent to high-volume autonomous operations that can create command dilemmas. To wit, commanders must retain positive control when thousands of kinetic effectors are simultaneously engaged, and operators must audit escalation when software updates occur in days rather than years. The framework forces explicit scoring of national-level approval burdens, ethical accountability, and capture risks; and confronts the true cost(s) of innovation that is responsibly pursued, with human judgment preserved at every layer. 

Table 2

Systems are evaluated against these twelve weighted criteria, organized into four groups, scored 1–7, and aggregated via Analytic Hierarchy Process (AHP) or Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) methodology. This produces auditable, repeatable outcomes in which no single factor dominates. The criteria are offered as an initial framework to stimulate discussion and refinement across the DoW Congress, industry, and allies.

The 10-Criteria Decision Matrix

The matrix integrates directly into existing DoW tools (e.g.  Adaptive Acquisition Framework gates, Program Objective Memorandum cycles, Joint All-Domain Command and Control data layers), automating scoring without reinventing bureaucracy. Requirements documents could specify “must achieve Level 5 or better under these criteria under projected near-peer-threat conditions,” with proposals evaluated against the same transparent rubric. Feedback on weighting, additional criteria, or implementation approaches from stakeholders is welcomed to strengthen this foundation for practical application.

Industrial Reality Check: When “Cheap” Ceases to Be Attritable

We argue that reducing “attritable” to a simple low-unit-cost imperative is dangerously seductive, but an incomplete approach. For example, in Ukraine, monthly first-person-view (FPV) and tactical drone consumption routinely exceeds 10,000 systems in sustained high-intensity operations, with production capacity scaling toward hundreds of thousands per month (see: The Hudson Institute: “Rise of Machine-Speed Adaptive Hyperware”). At near-peer competition scale (i.e.- as extrapolated through wargaming of protracted high-intensity scenarios against a near-peer industrial power such as China), demand could reach tens to hundreds of thousands of attritable platforms per month across domains (see: Center for Strategic and International Studies: “Is the United States Prepared for a War with China?” and The Diplomat: “The Drone Swarm Paradox in the Taiwan Strait”).

These calculations are sobering, even when limited to raw production costs alone (and excluding research and development, design iteration, operator training, maintenance infrastructure, storage, forward deployment, or software and cyber-security updates at combat tempo). Table 3 illustrates the escalation at a conservative $35,000 unit cost using the Low-Cost Unmanned Attack System (LUCAS) recently employed against Iranian targets as a baseline.

Table 3

These figures represent production costs only. The DoW must fully account for the logistics of storing hundreds of thousands of attritable systems in dispersed, hardened facilities, forward-deploying them under contested air and sea lanes and maintaining software integrity and cyber resilience when updates must occur at combat tempo across thousands of platforms simultaneously. These are not peripheral concerns; rather, they are the hidden multipliers that turn “cheap” mass into an industrial and operational liability when replenishment realism is ignored.

Allied Approaches and the Opportunity for U.S. Leadership

United States’ allies confront parallel challenges. The United Kingdom’s Defence Drone Strategy and Royal Air Force Autonomous Collaborative Platform Strategy tier uncrewed systems into disposable, attritable, and survivable categories. Both the European Defence Agency and NATO reference “low-cost attritable” platforms, yet these retain primarily size-based classifications. The February 2026 Low-Cost Effectors and Autonomous Platforms (LEAP) initiative launched by France, Germany, Italy, Poland, and the United Kingdom correctly emphasizes rapid production and joint procurement inspired by Ukraine but stops short of a multi-factor replenishment-realism spectrum.

The DoW ’s proposed framework closes the U.S. gap and positions the US military to lead harmonization through AUKUS Pillar 2 and NATO. By exporting a rigorous 1–7 scale and decision matrix, DoW can enable coalition-wide standards that distribute industrial surge demands and reinforce the administration’s expectation that allies and partners contribute meaningfully to shared deterrence burdens.

An Implementation Roadmap: From Concept to Policy

Obviously, the proposed framework is conceptual, yet we believe that it could be immediately actionable within the DoW’s five-year planning and funding cycles. Services and the Department could pilot the matrix on live Replicator-related programs during the next Program Objective Memorandum cycle, with full integration into requirements documents, sensitivity analyses, and congressional reporting required within 24 months. Outputs would be auditable and directly responsive to FY2026 NDAA direction. By year three, the refined definition and 10-criteria matrix could be embedded in updated acquisition guidance; by year five, allied harmonization through AUKUS Pillar 2 and NATO should produce standardized coalition surge protocols.

This roadmap operationalizes the 2025 National Security Strategy’s call for resilient supply chains and economic security as national security, while aligning with incentives to modernize defense acquisitions, strengthen the defense industrial base, and expand domestic critical-minerals capacity. The result would not be additional bureaucracy but instead, a disciplined approach to more efficient and effective acquisition reality.

Conclusion

Near-peer competition and consumption rates that are measured in the tens or hundreds of thousands per month demand intellectual honesty. The proposed attrition-tolerance framework, offering a clear definition, and operationalized through the proposed matrix, equips the DoW , defense industrial base, and allies with a shared, auditable capability that we believe is required for sustained advantage in great-power competition, and hardens industrial entities against  economic attrition that U.S. adversaries seek to impose. By moving to reinforce realism in terms, definitions and constructs in practice, the U.S.’s and its international partners can translate the 2025 National Security Strategy from vision to victory that is achievable through resilient, scalable, and ethically grounded instruments of national power capable of prevailing in the high-consumption conflicts that define the 21st century.

Disclaimer: The views expressed in this paper are those of the authors and do not reflect the official policy or position of the National Defense University, the Department of War or the United States government.