In 1983, Norman R. Augustine, a former defense aerospace business leader and Under Secretary of the Army, published his light satire on corporate management, Augustine’s Laws. The book draws strongly on examples and practices found in the defense sector at the time. It turns out not much has changed since the mid-1980s, and many of “Augustine’s Laws” are still valid in their forecasts of the future state of defense acquisition.
For example, in developing Law Number XVI, Augustine assembles the historical cost evolution of tactical aircraft. He extrapolates the trend and postulates the law: “In the year 2054, the entire defense budget will purchase just one aircraft. This aircraft will have to be shared by the Air Force and Navy 3-1/2 days each per week except for leap year, when it will be made available to the Marines for the extra day.”1 More than 30 years after Augustine published his book, the fundamental growth trend in the cost of tactical aircraft is holding true.
At first blush Law XVI may seem a little whimsical, but there is a lot of truth in its statement. We are experiencing exponential growth in the cost and complexity of developing tactical weapon systems and in the timeline of delivering them to the warfighter. As a result, it is becoming increasingly difficult to buy the needed quantities of weapons, train our warfighters to effectively use and maintain the associated systems, and transition needed technologies into the field.
This article is not focused on the effectiveness of our defense acquisition system, nor is it a comprehensive assessment of the cost of our modern weapon systems. Its purpose is to discuss emerging trends as we build our national defense capabilities, especially in tactical areas. The questions that need to be answered are: Can we sustain our current path of developing and deploying increasingly complex weapon systems? Are there other approaches that should be considered?
Cost, Complexity, and Schedule
One cause of cost growth is the increasing complexity of our weapon systems. Complexity can be measured in numerous ways. One simple metric is software lines of code, which for recent generations of tactical aircraft have soared. All our digital weapon systems have experienced similar growth.
Another metric of complexity is the number of sensors in a weapons system. An example of this metric’s growth is seen in the AIM-9 Sidewinder family of short-range air-to-air weapons. The AIM-9M has two on-board sensors, whereas the next-generation AIM-9X variant has 11 (counting its focal plane array as one sensor). The AIM-9X is also a digital weapon, so there is further complexity in the amount of software required to ensure all components function correctly.
The increasing guidance-system intricacy is resulting in an alarming rise in missile-guidance costs. A few years ago, a weapon-guidance set comprised roughly 10 to 20 percent of the weapon’s total cost. Today, with the increased complexity in more advanced missiles, that number is over 50 percent. Does the increased complexity provide a proportional gain in performance? Can the country afford this increase and, if so, is it the best way to ensure victory?
In particular, one key consequence of increased intricacy is that the complexity of guidance-system design increases interface requirements with the platform. Thus, the integration of these complex weapons with the platform has grown significantly due to the higher exchange of information between weapons and platforms.
It is important to note marginal increases in complexity may be worthwhile, but they need to be managed and deliberately employed. Unfortunately, elegance is not found in complexity; rather, it is found in simplicity. We increase complexity not only to leverage technology but also because we can.
Today, the rising cost and complexity of our tactical systems manifest themselves in lower inventories and higher maintenance costs that ultimately affect our warfighters’ proficiency. They cannot afford to train adequately because these more complex systems require more training than simpler systems.
A 1980 government report on highly sophisticated weapon systems concluded:
. . . many of the weapon systems developed by the United States today are viewed as being too technologically complex to permit a reasonable degree of confidence that they will work properly when needed. Consequently, the Nation’s ability to be sufficiently prepared to sustain itself in a major war is of serious concern. Several problems that have resulted from the Department of Defense’s (DOD) acquisitions during the 1970s include: 1) few weapons are available due to high unit cost; 2) weapons have reliability, availability, and maintainability problems; 3) small annual procurement quantities are uneconomical; 4) high operating costs tax training resources; and (5) complexity and sophistication aggravate personnel problems.2
These are not new problems, nor are they problems that we have been able to manage since that report was published. Rather, the situation has gotten worse.
No physical system can continue to grow exponentially forever. Sooner or later, the system will break. The ever-increasing complexity of our weapon systems is no different. In fact, we are beginning to see telltale signs of such breakage: programs being canceled or accepted at significantly reduced performance than specified, increasing frequency of performance failure, loss of availability due to reliability problems, and higher cost of system maintenance. The cancellation of the Army’s Future Combat System is a good example of a system in which the desired intricacy was so high that it could not work without significantly more money and time.
Augustine’s Law XVI is a tongue-in-cheek prediction of a pending breakage. The exponential growth in complexity, cost, and schedule we are experiencing in our national defense elements cannot be sustained. The Soviet Union is a prime example of the negative effects of such an increase, evidenced in the aftermath of its response to the Reagan defense buildup. Its economy could not sustain such growth and the result was a fundamental restructuring of its system. Are we on a similar path?
The Current Approach
The United States basically plans for two types of warfare: 1) major warfare against a near-peer nation state, in which strategic weapons would be used, and 2) regional warfare, much like what we have experienced since World War II and most recently in the Middle East, in which tactical weapons would be most often used. Today, we typically use the same approach for developing both our strategic and tactical weapons. Is this necessary or wise?
One of our main approaches to maintaining our national security is to leverage technology by developing and deploying increasingly sophisticated weaponry as threats evolve. In particular, our goal has been to quickly transition leveraged technology to new weaponry to stay ahead of threats. However, we do not transition technology fast enough to keep up with rapidly evolving threats. We routinely overstate the requirements for a new weapon system, understate the readiness of the technology to be employed, and are overly optimistic regarding the cost and time to develop and deploy these new systems. As a result of this optimism, the development and deployment of such systems either does not happen or takes longer than planned and costs more, and the technology employed is often obsolete by the time the new system is fielded.
What Can Be Done?
These two types of warfare demand different approaches and tradeoffs of cost, complexity, and schedule for developing and deploying the weapon systems involved. For strategic systems, the tradeoffs between these parameters may lean toward higher cost and complexity, since the stakes of major warfare are very serious. The destructive power of these systems is extremely high, while the production quantities tend to be relatively low. For regional conflict (which includes conventional warfare, asymmetric warfare, electronic warfare, and guerrilla warfare), the necessary quantities of weapon systems tend to be much larger, since the probability, frequency, and duration of regional combat is higher. The tactical weapon systems needed should be affordable so inventories can be bolstered, and be able to respond to a broad range of threats that tend to adapt very quickly. The cost, complexity, and schedule associated with these weapon systems should be responsive to these needs.
The threats we may face in the future are significant and challenging. They could be extremely sophisticated or very rudimentary, fairly dense or widely dispersed. Our tactical weapon systems need to be able to address all of these risks. Today, we seem to think that we can maintain an advantage solely through developing increasingly sophisticated technology. Relying on silver, or maybe platinum, bullets does not constitute a sound or affordable strategy. Such weapons may seem attractive, but they are costly and can be countered. This does not mean we should quit trying to modernize our tactical weaponry or improve its capabilities through technology. We just need to use a different approach.
Warfighting is about projecting power, and obtaining and effectively using information. A typical kill chain used in prosecuting a threat encompasses six basic steps: finding the target, fixing or accurately locating it, tracking it as it maneuvers, identifying and prioritizing it, engaging it, and, finally, assessing the engagement. Five of these steps involve information; one step, engagement, is about power projection. Today, we use a plethora of sensors (i.e., electronic, electromechanical, and human) to find, fix, track, identify, and assess various targets. We develop capable platforms to house these sensors and survey the geographic areas and targets of our interests, thereby generating the information needed for our decision makers to act.
Once we decide to attack a target, we essentially discard the obtained information and have the weapon regenerate the same data on its run-in to the target. Regenerating the information is the function of the missile’s guidance system. To further complicate the task, we ask the missile guidance system to regenerate information in all kinds of weather and environments, e.g., fog, smoke obscuration, the presence of decoys, electronic-warfare effects, and bad weather. All of these these factors combine to make the guidance set increasingly more complex. The cost of missile guidance is rising at an alarming rate, and in the end, the guidance system is destroyed. We could significantly reduce the complexity and cost of our missiles if we would reuse the information already generated by other elements of the kill chain.
To reuse the information already generated by our platforms and sensors, it needs to be time-stamped, secured, stored, and transported to all the destinations where it will be used, including the missile. The whole idea is to stop concentrating the needed complexity into single systems. By reusing information and distributing the complexity across the kill chain, the individual systems that make up the overall system can be simpler, less sophisticated, less costly to produce and maintain, and consequently, produced in larger numbers. Although a complex kill chain will require increased training, higher inventories that are cheaper to maintain will give our warfighter the ability to train more thoroughly to maintain proficiency. Higher inventories also will allow multifaceted strikes that can flood enemy defenses.
Interoperability is Key
The success of this approach centers on a distributed interoperable capability employing simpler agents (e.g., airplanes, weapons, sensors, and ships) that act together. Networking these agents so they interoperate to provide the functions needed allow military planners to scale the magnitude of the network (i.e., the number and type of agents involved) to the tasks at hand. The technology is available today to enable this approach. However, we will have to learn to fight differently. We will need new tactics and doctrine. Success depends on how we implement technology, our technological capabilities, and a willingness to operate in a network that relies on other elements to create the desired outcomes.
This approach promotes the agility, flexibility, adaptability, and asymmetry needed to address all kinds of potential threats: large or small, sophisticated or rudimentary, or single or multifaceted. This strategy also would enable us to transition and insert technology at a significantly faster rate, and at much lower cost and risk. Continually evolving our capability will allow us to maintain dominance. This “Lego” strategy of using many simple pieces, interoperating to build sophisticated outcomes, provides the agility to preplan or react to any potential adversary.
Networking simple agents that act in unison to create capability is not a new idea; in fact, it is a strategy used often in nature. As Sun Tzu professed in The Art of War, “There are not more than five musical notes, yet the combinations of these five give rise to more melodies than can ever be heard. There are not more than five primary colors, yet in combination they produce more hues than can ever been seen. There are not more than five cardinal tastes, yet combinations of them yield more flavors than can ever be tasted.”3 Simple agents acting together can create more capabilities than will ever be needed to address any potential enemy.
On 27 January 2015, the Naval Air Warfare Center Weapons Division demonstrated this approach using a Tomahawk cruise missile fired from the USS Kidd (DDG-100) near San Nicolas Island in California, targeting a moving ship. The demonstrated antiship capability was realized with assets already fielded for other purposes but assembled differently to create an alternate end result. A P-3 patrol aircraft equipped with a littoral-surveillance radar system served as a targeting source, tracking a surface-ship target from a distance. Based on the target’s position and estimated velocity, the Tomahawk’s aimpoint was continuously updated via the Tomahawk Strike Network. The updated aimpoints were used to make in-flight missile course corrections toward the moving target. The test demonstrated the ability to send a long-range weapon on a trajectory that could be updated in real time to put it on a collision course with a moving target at sea.
Every capability has its vulnerabilities. The naysayers will criticize this approach as too vulnerable, perhaps countering with “Break the network and it falls apart,” or “I must trust others to provide me needed information.” While true and crucial to address, in reality, these vulnerabilities are already emerging. The way we fight today and the way we will fight tomorrow includes networks of systems and people working together and communicating with one another to perform the mission at hand. We will need to seriously address how we connect these elements together. Connections need to be dependable, agile, available, secure, adaptive, and reconfigurable so the information is trustworthy. Information underpins every aspect of warfare; no matter how we fight, it is critical to transport it, secure it, know where and when it was collected, and make it trustworthy and available to those who need it.
A Way Forward
First, we must change the way we write requirements for our warfighting systems. The requirements that drive complexity and cost must be carefully approached. For example, the requirement for a high single-shot probability-of-kill seems at first glance like a way to save money in production by “needing” a small inventory of weapons. This one requirement, however, drives complexity, developmental costs and schedule, high unit-production costs, and the need for high survivability. The fact is, in the heat of battle, warfighters do not plan for or use merely a single weapon for killing a target. They use many, “just to be sure!” They do not want to risk the loss of time, personnel, and equipment by having to restrike the target.
While the requirement for a high single-shot probability-of-kill weapon may appear to be pragmatic and expedient, in reality it is a wolf in sheep’s clothing. The requirements for our warfighting systems need to be carefully thought through. They need to drive toward simplicity, ease of integration and interoperation (i.e., “plug and play”) with other systems, and leverage information that exists. Summarily, simple systems that are integrated and interoperate enable higher inventories, more comprehensive training, multiplicity of capabilities, and innovation by our warfighters on the battlefield.
Second, the government must take control of our warfighting capability. This control is something we abrogated when we restructured our defense industrial base at the end of the Cold War. Our defense industry is very good at developing and building our warfighting systems. It is not, however, incentivized to make those systems interoperate easily and readily to yield capability. In fact, we “require” development of stand-alone systems and not interoperability on a broad scale. The government needs to develop standards and interface reference designs that govern integration and interoperation.
We need to use the telecommunication industry as a model for such standards. Telecommunications standards enable a broad set of developers (hardware and software) to participate in the industry at large. These standards are developed and governed by an entity bigger than any individual company, an entity that has stewardship over the whole of the industry itself. For our national security systems, that entity must be the government.
Third, the government must return to the practice of buying the technical data that defines our warfighting systems and must disallow proprietary systems. By their very nature, proprietary systems do not interoperate. Technical data will provide government engineers the necessary information to ensure systems are designed to interoperate, and will help verify that the systems perform as desired.
Fourth, we need to invest in information-transport technology and transport media that are robust and secure.
Lastly, we can implement this approach with very little new investment. We do not have to wait and bear the cost of developing and fielding a new set of weapon systems to enable this integration and interoperability strategy. We have significant capability available today. We can begin immediately by developing the required standards and a set of reference designs that dictate how we want the interfaces among systems to be implemented. We can start using our current systems in the manner we envision future systems to interoperate by conducting experiments with fielded systems in the fleet. The outcomes would be new concepts of operations and procedures, new system interoperability requirements, and prototypes for new warfighting systems we need. This approach mines the latent capabilities of currently fielded systems integrated to provide stop-gap measures that address emerging threats. It leverages and employs the resources we already have in different ways; it is a deliberate crawl-walk-run strategy toward building a new paradigm for warfighting equipment. The results from these experiments will help us write the requirements for future systems.
In 1983, Augustine raised a red flag about the unchecked cost growth of our warfighting systems. Since then, we have devised no strategy to counter this trend. In fact, we have exacerbated the issue by requiring increased complexity and performance in our various warfighting systems. This approach cannot continue without significant consequences. Rather than demanding that our systems be able to fight alone, we can ensure they actively interoperate and share information, allowing us to field simpler systems that are cheaper to procure and maintain.
Distributing the necessary complexity across all elements of a kill chain, rather than placing it singularly in the nodes of the network, we can reduce the ownership costs of our warfighting capability, insert technology in a faster and lower-risk manner, and scale the specific response to the challenge at hand. We will be able to realize significantly increased inventories. The resulting kill chains may be more complicated, but lower cost and increased inventories will allow our warfighters to train with real systems rather than surrogates to maintain adequate proficiency. Emphasis on simpler systems interoperating to share information is a major shift, but it will result in reduced cost; faster employment of technology; increased innovation, flexibility, and agility to address emerging challenges, and improved warfighter teamwork and proficiency.
1. Norman R. Augustine, Augustine’s Laws (Reston, VA: American Institute of Aeronautics and Astronautics Inc., 1997 republication of 1983 edition).
2. U.S. Government Accountability Office, “Implications of Highly Sophisticated Weapon Systems on Military Capabilities” (PSAD-80-61), 30 June 1980, www.gao.gov/products/PSAD-80-61.
3. Sun Tzu, The Art of War, trans. James Clavell (New York: Delacorte Press, 1983).
Mr. O’Neil began working at what is now known as Naval Air Warfare Center Weapons Division in 1972 and served as the NAWCWD executive director for more than eight years before recently retiring. He was formerly appointed to the Senior Executive Service as head of NAVAIR’s Weapons/Targets Department. He has a master’s degree in management from MIT, a master’s degree in mechanical engineering from the University of Southern California, and a bachelor’s degree in mechanical engineering from Seattle University.