WASHINGTON – The U.S. Defense Department’s space acquisition programs contain too much development work that is better suited for its science and technology laboratories, according to a recent government audit report.

Developing new technology as part of the acquisition process has been a major factor in cost overruns and schedule delays on military satellite programs, according to a Government Accountability Office (GAO) review of space science and technology work.

The report, “Technology Development: New [Defense Department] Space Science and Technology Strategy Provides Basis for Optimizing Investments, but Future Versions Need to be More Robust,” lauded the Pentagon for improving its approach to space science and technology, but highlighted a few shortcomings.

The report was called for in the National Defense Authorization Act for 2004 , and was submitted to the chairmen and ranking members of the Senate and House Armed Services committees Jan. 28.

The legislation laid out nine requirements to govern Pentagon space science and technology work. The GAO study concluded that the Pentagon has met four of those: identifying short- and long-term goals; establishing a process for achieving them; developing that strategy in consultation with the relevant military organizations ; and setting a process for reviewing its space science and technology strategy.

The congressional watchdog agency found that the Pentagon is on its way to meeting the other five: conducting annual reviews of the space science and technology strategy; making the strategy available to Capitol Hill; including the review in its National Security Space Plan; sharing the strategy with the relevant Pentagon organizations for budget planning purposes; and having Pentagon labs submit their space program budgets and schedules to the director of defense research and engineering as well as the undersecretary of the U.S. Air Force.

But the GAO also identified several problem areas.

Those included difficulty coordinating space science and technology efforts funded by the Air Force, the National Reconnaissance Office, Missile Defense Agency, Defense Advanced Research Projects Agency and other organizations. The report also found that while launch costs have risen, funding available to launch space experiments declined from $62 million in 1990 to $39 million in 2004.

The auditors also found that the Pentagon generally invests 80 percent of its research and development funding in acquisition programs, with the remaining 20 percent spent in science and technology labs.

Despite the heavier investment in acquisition-specific research and development, which is spread over far fewer efforts than in the science and technology labs , competition for dollars often leads program officials to latch onto immature technology that offers the promise of significant performance gains, the report said. When these technologies take longer to perfect than expected, it creates a ripple effect that delays and drives up the cost of entire program , the GAO said.

“Events such as test ‘failures,’ new discoveries, and time spent in attaining knowledge are considered normal in this [the laboratory] environment, while they are seen as a negative event in an acquisition program,” the report said. “Moreover, separating technology development and product development enables organizations to align customer expectations with resources, and therefore minimize problems that could hurt a program in its design and production phases.”

The Pentagon appears to be making an effort to address this issue for most systems, but space acquisition is covered by a separate policy that allows for significant technology development during acquisition efforts.

The space acquisition policy is under review and officials involved are re-examining how long technology development should be allowed to continue within an acquisition program, the report noted .

Paul Kaminski, a former Pentagon acquisition chief, said the Defense Department might be able to avoid much of the turmoil encountered on nearly all of its satellite efforts by allowing its science and technology labs to perform the initial technology development work .

These laboratories take a slower, steadier pace than is possible in the midst of an acquisition program , Kaminski said. Snags are part and parcel of technology development, and during acquisition they can put everything on hold, driving up the cost of the program, he said. “We should be preparing for a marathon, not a sprint,” Kaminski said.

James Lewis, director of technology policy at the Center for Strategic and International Studies, a think tank here, said the efficacy of developing technology during acquisition depends on the program. In some cases, acquisition programs are the best place for innovation because of their direct connection to an objective system, he said.

Science and technology labs are the best avenues for finding general solutions to problems, like identifying laser-optical links as a method for increasing the speed with which satellites can transmit information, Lewis said.

Once a solution like laser-optical links is identified, the development work can be handled within an acquisition effort by program officials with more experience building operational satellites, he said.