Do you know "Cheop's Law"? Named for the Pharaoh who built the great pyramid, and postulated by the author Robert Heinlein, it runs:"Nothing ever gets built on schedule or within budget." Anyone who questions the wisdom of this maxim should examine the Defense budgets of the world's democracies, apart from the average home remodeling project.
In spite of decades of experience with this sad reality, there is a large and influential network in Washington dedicated to acting astonished every time a military or quasi-military program exceeds its budget. It works like this - the General Accountability Office (GAO) issues a report and the media acts "shocked, shocked." Congress sometimes holds hearings and passes laws and regulations to prevent this from ever happening again, and lawyers in the Pentagon and at the big aerospace firms smile and enjoy their job security.
The continued failure of most new large aerospace and defense projects to meet their cost, schedule and performance targets cannot be wholly blamed on many people's favorite targets: greedy contractors and warmongering Generals, nor on the conservatives preferred whipping boys of excessive regulations and the anti-defense liberals in Congress. The important part of the problem seems to be centered in the way the universities train America's engineers and the procedures, and processes that industry has been using for more than fifty years.
Ever since the early 1960s, when Secretary of Defense Robert MacNamara reorganized the Defense Department's procurement system on so-called "rational" lines, the US Military's method of buying major weapons and support systems has been an expensive mess. In spite of this, the process repeatedly produced the world-beating technology that eventually helped convince the Soviets that they could never win the Cold War; and since 1991, has given America's troops the tools they need to defeat any enemy on any battlefield. All this for an average of less than 6% of GDP would seem quite a bargain, especially compared to the alternatives of defeat or nuclear war.
Today it is said that a combination of bureaucratic inertia, legal and regulatory overkill, lack of procurement expertise and the need for very high quality and very effective weapons drives the cost of maintaining a modern military establishment to ever higher and ever more unaffordable levels. Every few years someone in the Pentagon or Congress tries to reform the system, sometimes by changing or tightening up the regulations, sometimes by letting the defense Industry operate with less supervision -- and most recently, by increasing the number of trained procurement specialists keeping an ever closer eye on the firms that design and build America's military hardware.
A 1990 study of the problem by Kenneth Adelman and Norman Augustine explains that, "... Senator Sam Nunn, among the nation's most respected experts in national security matters observed that many of today's procurement problems 'were yesterday's procurement solutions.'" Earlier, in 1983, Augustine wrote Law number 36: "The thickness of the proposal needed to win a multimillion dollar contract is about one millimeter per million dollars. If all the proposals conforming to this standard were piled on top of each other at the bottom of the Grand Canyon, it would probably be a good thing. " This is nothing new, Ancient China had a similar problem: according to one sinologist, "It was a regime of paperwork and harassment {paperasseries et tracassseries } , endless paper work and endless harassment."
America, like most of its allies, is investing a constantly shrinking share of its national wealth in its armed forces. As of 2010, it was estimated that the US spends about 4.5% of its GDP on defense. In contrast, during the Reagan-era build-up, the Defense Department and the other parts of the defense budget represented about 6.5% of GDP in 1985, and 6.4% in 1987. In neither of those years was the US actively fighting a war. In the late 1950s when the US military was mostly manned by low paid conscripts, the government spent roughly 10% of GDP on the Defense budget.
The US should be getting better results for the money it spends. The quality of an F-22 air superiority fighter , for example, is not in question, but if the President and Congress decide that we can only afford 187 of them, compared to a certified need for 380, then something is terribly wrong. The same problem of excessive costs leading to a severely curtailed procurement, afflicted the B-2 bomber: only 21 were bought when the air force needed about 120. Today, the F-35 Joint Strike Fighter is in danger of being cancelled or curtailed due to an estimated overall 65% cost increase since 2002.
&nbs p; In the 1950s, nuclear weapons supposedly gave America "More bang for the buck." In the 1990s, it was claimed that precision-guided weapons would, theoretically, reduce the costs of fighting a war as fewer munitions would be needed to destroy any single target. Yet the increase in overall military costs, per airplane or per ship, show no sign of slowing down: high performance technology comes with high costs, This is equally true of civilian programs that have nothing to do with the US Government's procurement system. In recent years, both Airbus and Boeing have run into big and expensive problems with their new A-380 and 787 civil airliner programs. Cheops goes after soldiers and bureaucrats as well as civilians.
Perhaps it may be time to look beyond the old debates about regulation and authority? The arguments over whether the Pentagon needs more lawyers and bureaucrats, or more scientists and engineers, keeping a close eye on Boeing, Lockheed Martin, General Dynamics and the other large aerospace companies, may not be relevant. Instead the cause of the problem may lie in the interaction between today's digital- or cyber-technology and the old discipline known as 'Systems Engineering."
One indication that the problem is possibly not the exclusive fault of the military procurement system is the fact that many other government aerospace projects are running into problems that closely resemble those of the military -- for example, the NASA James Webb Space Telescope (JWST) . This long running program -- to build a high performance successor to the famous Hubble Space Telescope (HST) -- is in deep trouble: The latest report says that it will now cost at least $1.5 billion more than the previously estimated $5 billion to finish. Other large NASA science programs are facing similar problems.
Outside of government, Boeing's advanced 787 Dreamliner program has just announced yet another delay; the company will be lucky to deliver the first aircraft to its leading customer by the end of this year -- aside from having to conform to long-established US and International safety regulations The 787 does not have to conform to the kinds of government procurement regulations that have been blamed for cost overruns and delays in military projects. In Europe, the giant Airbus A-380 faced similar costly difficulties. Could it be that the world is not facing a set of problems that requires legal, bureaucratic or procedural reforms, but something far more profound -- such as that the engineering methods and procedures that have been used by the Aerospace industry are failing to meet customer expectations?
In a March 2007 speech at Purdue University, former NASA Administrator Michael Griffin pointed out that high-performance, complex systems usually do not "accomplish the tasks for which they were explicitly designed. Complex systems typically fail because of the unintended consequences of their design, the things they do that were not intended to be done. The Second Law of Thermodynamics (describing entropy) is sufficient to guarantee that most of those things will be harmful. " So any program the budget and schedule of which was based on "known' factors, experience and 10-20% "fudge factor" will, based on current engineering practices, exceed its budget and schedule by at least 100%. The more complex the system, and the more difficult the task it is required to accomplish, the more certain we can be that the initial cost and schedule estimates will be massively wrong.
A corollary to this is that it is easier to build improved versions of existing systems. Aircraft ,such the the C-130 Hercules transport and the Huey and Chinook helicopters, have been in production since the 1950s. They have been upgraded and heavily modified, but they are recognizably still the same flying machines they were back when Eisenhower was in the White House. The lesson may be that good designs tend to last, and also that the quality of these designs is known, so building new versions does not require 100% new hardware and software.
From the 1950s to the 1980s, large aerospace firms were masters of something called "Systems Engineering," which involves "decomposing a design into separable elements, characterizing the intended relationships between them, and verifying that the system is built and operates as intended." This method helped get America to the Moon, and also helped make the US aerospace industry the envy of the world. This is no longer the situation, Boeing has lost its supremacy to Airbus, and the rest of the industry faces a nightmare of overregulation, shrinking budgets and a lack of scientific and engineering talent.
Lockheed Martin's VP for engineering was quoted in in Aviation Week last year as saying: "Historically, systems engineering has been successful in bringing order to the development of of systems as they have become increasingly complicated. But there is a big difference between complicated and complex" ... "Complicate d is decomposable. which is what systems engineering is based on. Complex systems are no longer strictly decomposable, and systems engineering has to adapt." In the same article Michael Griffin agreed, pointing out that, "There are cases where everything thought necessary was done, and somehow something went wrong anyway." Cheops is a hard taskmaster.
One attempt, at least temporarily, to get away from the business-as-usual approach is the Defense Advanced Research Projects Agency 's (DARPA) Advanced Vehicle Make (AVM) program -- in particular its META sub program. The goal of META is ".. to make dramatic impact on the existing systems engineering, integration and testing process for defense systems." The most important and most difficult aspect of this ambitious program is to get away from the old design-build-test, redesign-rebuild-retest process that is the cause of so many of the cost overruns and delays that plague most contemporary attempts to produce high performance systems. DARPA hopes that the AVM program will result in a new manufacturing process that will be more like the ones used by the semiconductor industry, and less like the trial-and-error techniques used to make a handmade suit. If the aerospace and defense industry can learn to design and build new systems that do not require the expensive and time-consuming testing and modifications that are typical of today's projects, it will have gone a long long way toward making cost overruns and project delays problems of the past.
If Congress, the leaders in the Pentagon and the executive branch cannot reform the way that they interact with industry, and vice versa, we will continue to see expensive programs cancelled and billions wasted, while America's technological superiority disappears down the drain.
