March 11-17, 2002

Death spiral, downward spiral, spiraling costs, spiraling out of control — all have been applied to U.S. weapon programs delayed by technological barriers, developmental challenges or skyrocketing costs. Now comes spiral development, touted by Pentagon acquisition chief Edward "Pete" Aldridge as the best way to speed development and deployment of weapons and save money.

Surely, he will forgive those of us who are just a bit skeptical.

Acquisition reform has a long and, sadly, unimpressive history. During the past two decades or so, the time between program launch and fielding of weapon systems has, if anything, increased; the current average is 10 to 15 years.

Spiral development, in fact, is a term long used by the commercial software industry for the process of rapidly providing new products, and fixing bugs that crop up through technical updates and patches.

The Pentagon has toyed with the concept for years. But details about its latest incarnation, and how it will work when applied to major weapon systems, remain rather vague. Aldridge says the guiding principles include:

	Doing away with fixed, threat-based military requirements in favor of so-called capabilities standards set by the developers, industry and service users.
	Relying more heavily on commercial technologies.
	Building initial versions with currently available technology
	Rapidly fielding prototypes to get user feedback.
	Gradually improving operation as new technology becomes available.

The goal, he says, is to cut time between program initiation and production to five to seven years.

Obviously, service budgets and soldiers in the field would benefit from less costly, more rapid weapon development programs. On the other hand, there are serious risks involved with buying and fielding equipment that may not be in true working order.

Even under the old acquisition system, with its intricate web of checks and balances, there were strong political and budgetary pressures on service program managers to move ahead with development and production even if performance and testing fell far short of requirements. The V-22 Osprey debacle comes to mind.

In addition, it is unclear whether commercial software development and production practices can safely be applied to highly complex weapon projects.

Moreover, commercial firms have strong financial incentives to get it right. If consumers do not find one company's product worthy, they can switch to a competitor. If the product is truly defective, consumers can sue. Neither of these remedies is available to taxpayers or soldiers if a Pentagon weapon program flops.

Even if Microsoft's latest version of Windows crashes office computers on a regular basis, nothing but a few hours (maybe days) of company productivity is lost. If the Air Force's Joint Strike Fighter crashes multiple times, that is an entirely different story.

Even more troubling is that the technically ambitious missile defense effort of President George W. Bush is to be one of the first and highest priority programs to demonstrate spiral development. The Pentagon just overhauled the missile defense program, and the Pentagon agency managing it, in a manner that many members of Congress and watchdog organizations fear will reduce oversight dramatically. Spiral development is one, but not the only, culprit.

For example, the new Missile Defense Agency (MDA) has abolished all former Operational Requirements Documents for missile defense developments — documents that previously specified such requirements as range, speed and lethality of each proposed interceptor system.

Instead, using spiral development practices, system performance goals will be based not on countering specific threats through specific requirements, but on determining what capabilities might be needed to counter any enemy capabilities. These capabilities standards, moreover, might be only 70 percent to 80 percent met prior to fielding, according to Pentagon officials.

The MDA also will be free of many of the usual Pentagon reporting requirements, including those about cost growth. It also will be able to avoid running much of its testing program through the Pentagon's independent Office of Test and Evaluation.

Finally, the MDA program budget has been reorganized to blur the lines between discrete technology approaches or systems, making spending even harder to track. According to the Pentagon, the new structure is designed to create a better-integrated, layered system based on three major program segments.

Lt. Gen. Ronald Kadish, MDA director, told the House Armed Services Committee Feb. 27 that the Pentagon is requesting $38.7 billion in missile defense research and development between 2003 and 2007. Since the Pentagon has yet to lay out a proposed deployment architecture, it is extremely difficult to predict production and operation costs.

But the Congressional Budget Office, in a Jan. 31 report to Tom Daschle, Senate majority leader, estimated the cost of deployment and operations through 2015 of a simple, three-site ground-based midcourse system — only one of the multiple ground-, sea-, air- and space-based layers in the new MDA concept — at between $56 billion and $68 billion.

Surely it behooves the Pentagon and Congress to ensure that such spending is adequately tracked, and that systems fielded actually work as advertised.

Rep. Martin Meehan, D-Mass., hit the nail on the head in his remarks to the House Armed Services Committee hearing: "The MDA is using requirements that are not built on a threat. ... I'm concerned that without a threat to work against and with no accountability ... we're freeing the MDA to embark on a Rube Goldberg, multimillion dollar, gold- plated science fiction project."

He and the rest of us have a right to be concerned.

Theresa Hitchens is vice president of the Center for Defense Information, an independent think tank in Washington.

 
First published in Defense News March 11-17, 2002.

By Theresa Hitchens
CDI Vice President
thitchens@cdi.org

BACK TO THE TOP    MISSILE DEFENSE HOME    CDI HOME