A review of the Pentagon's missile defense project by a panel of independent experts found no fatal flaw but questioned whether it can be completed by 2005 as scheduled, officials said Monday.

The Pentagon set 2005 as its goal for completing the project because it coincides with the CIA's estimate of when North Korean is likely to have a long-range ballistic missile capable of striking American territory.

President Clinton has not yet given the go-ahead to build a national missile defense, but the Pentagon is proceeding under a congressional requirement to develop the required technologies as quickly as it can.

Some critics assert that the technology for shooting down missiles in space is fatally flawed because a hostile nation could fool U.S. interceptor rockets by using decoys and other measures. The independent review panel advised the Pentagon to put more emphasis on overcoming such decoys, officials said.

The panel of former military officers and weapons scientists, headed by retired Air Force Gen. Larry Welch, presented its conclusions in a classified report to Defense Secretary William Cohen, who is preparing to make a recommendation to Clinton in August on whether to move forward with deployment.

A newly released report by the General Accounting Office, the investigative arm of Congress, makes similar points. The GAO report notes that key decisions about moving ahead with missile defense are scheduled to be made this year before the actual rocket used to boost the missile interceptor into space is fully tested.

The booster to be used in the next missile intercept test, in early July, is a stand-in for the three-stage rocket that would be part of an actual missile defense system.

Air Force Lt. Col. Rick Lehner, spokesman for the Ballistic Missile Defense Organization that is running the project, said his office is confident the actual booster will be ready for use in flight tests in 2001.

"It shouldn't be a problem," Lehner said.

Besides questions about feasibility, some critics question the cost of missile defense. The Pentagon estimates that a system with 100 missile interceptors would cost $36 billion over a 20-year operational life.

Gen. Michael Ryan, the Air Force chief of staff, told reporters Monday that he believes missile defense "can be affordable," but he added that "We don't have great definition on how much (it) is going to cost."

Despite cautioning that the project's timetable may be overly ambitious, the Welch panel concluded that the missile defense system is technically feasible and eventually could be made to work reliably, according to a senior defense official who has seen the report. The official spoke on condition of anonymity.

The reviewers saw "no reason to change the schedule at present," the official said.

The conclusions, first reported Sunday by The Washington Post, carry considerable weight with Pentagon decision-makers. The Welch panel, which is granted access to secret military data, has performed periodic assessments of the program; in response to last fall's report the Pentagon asked Congress for additional money for testing and extra equipment that the Welch panel had flagged as shortcomings.

Kenneth Bacon, spokesman for Cohen, said he could not comment on the latest report because it is classified. But he said it does not raise issues that necessarily jeopardize the 2005 target date for deployment.

"We regard this as an encouraging report because it says we're on the right technical path to meet the planned 2005 deployment date," even though everything will have to fall neatly into place to make that date, Bacon said.

Cohen and other Pentagon officials have publicly acknowledged a high risk of failing to meet the 2005 deployment target date. They say they will adjust the schedule if necessary, based on such factors as flight test results, arms control issues, cost and intelligence estimates of the evolving ballistic missile threat.

The Welch panel recommends that the Pentagon put the system — a combination of powerful radars, ground-based missile interceptors and high-speed computers — through more demanding tests than presently planned. It says tests so far have been "less realistic than they should be," the official said.

The degree of realism in the tests is, in part, limited by restrictions on how the Pentagon uses its missile test range in the central Pacific Ocean, the official said.