The Pentagon's effort to develop an antimissile system to defend the United States remains plagued by inadequate testing, spare parts shortages and management lapses, according to an independent panel appointed by the Defense Department.

In a stinging 40-page report, the panel warned that recent delays in testing and development have "compressed" the program's schedule against politically imposed deadlines. If further delays arise, the panel advised, plans for President Clinton to decide by next summer whether to start building the system should be postponed.

The critical assessment followed a report by the same outside group in early 1998 registering some of the same concerns and cautioning against a "rush to failure" in the nation's renewed multibillion-dollar drive to erect a limited shield against ballistic missile attacks, a legacy of President Ronald Reagan's much grander "Star Wars" vision. While the Pentagon has taken steps since last year to extend the earliest deployment date from 2003 to 2005, schedule additional tests and hire the Boeing Co. to coordinate development efforts, panel members said the program remains at "high risk" of failure.

The report faulted government and contract officials for exhibiting "a legacy of over-optimism" about their ability to develop a reliable interceptor that can soar into the sky and ram incoming enemy warheads. Although written before the first successful intercept test last month of the latest "kill vehicle" prototype, the report noted that the Pentagon's history of chasing missiles in space is littered with many more misses than hits.

In a program that ranks as the Pentagon's most challenging development effort--fraught as it is with technological hurdles, political controversy and fierce opposition from Russia and China--the panel also found troublesome management gaps.

"Instead of unusual clarity, there is unusual fragmentation and confusion about authority and responsibility," said the report, which Pentagon officials quietly sent to Congress last week after taking more than two months to review it.

The critique comes in the face of strong Republican support for a national antimissile system and begrudging acknowledgment this year by the Clinton administration and congressional Democrats that a limited system--to protect against a few missiles at any one time--may be needed sooner rather than later to guard against the growing threat from such "rogue nations" as North Korea and Iran.

GOP proponents were quick to draw what encouragement they could from the panel's blunt assessment. Sen. Thad Cochran (R-Miss.)--who pushed the bill enacted earlier this year requiring the deployment of an antimissile system as soon as "technologically possible"--issued a statement saying that, while the panel has rated the program's inherent risks as high, they did not appear "unacceptably" high.

He said any defense program subjected to such close scrutiny "will face some criticism." He also argued that the risks were outweighed by the urgency of getting a system in place to defend against potential attacks from nations such as North Korea, whose test of a three-stage Taepo Dong I missile in August 1998 caught the U.S. intelligence community by surprise.

"We don't have the luxury of time," Cochran said. "Because of the threat, we have no choice but to accept a high-risk program."

But critics of the antimissile program saw in the panel's findings fresh cause to urge a go-slow approach.

"In our view, delaying the program until there's more certainty of success is a reasonable course," said Steve Young, deputy director of the Coalition to Reduce Nuclear Dangers, a nonprofit group of 17 arms control organizations. "This is something you don't want to get wrong, because the consequences, if it fails, can be disastrous."

At the Pentagon, the Ballistic Missile Defense Organization, which coordinates the government's various antimissile programs, issued a two-page statement concurring with most of the panel's recommendations for additional tests, more hardware and tighter oversight. But it was silent on the proposal to postpone the president's review if new delays arise.

In addition to determining whether the system can work, senior administration officials have said several other factors will influence their recommendation whether to proceed. These factors include the projected costs of fielding the system, intelligence assessments of enemy threats and Russia's willingness to negotiate changes in the 1972 Anti-Ballistic Missile Treaty.

Under the system envisioned by the Pentagon, the launch of an enemy missile would first be detected by space-based military satellites, then tracked by ground-based early warning and X-band radars. Interceptor missiles based in Alaska would be fired to home in on the incoming warhead and collide with it at supersonic speeds in what the Pentagon calls a "hit to kill."

But this approach to missile defense remains fraught with technical problems. While a prototype did score a successful intercept last month over the central Pacific, no tests yet have attempted to integrate the entire system: interceptors, radars and controlling computer networks. Only two such tests are scheduled between now and the moment next summer when Clinton is slated to decide whether to begin building the system at an estimated cost of more than $10 billion for the first phase alone. Actual operational versions of the interceptor and booster, as opposed to prototypes and surrogates, are not due for several years.

Consequently, instead of confronting Clinton with a "deployment readiness" decision next summer, the panel suggested that the review be converted into a "development feasibility" assessment, effectively buying time.

The 12-member panel is the most experienced collection of civilian and retired military officers to have studied the antimissile effort, defense officials said. Headed by Larry Welch, a retired four-star general and former Air Force chief of staff, the panel includes specialists who earlier oversaw the development of missile, aeronautical and naval programs while working either at the Pentagon or for major defense contractors.

Given the tough technical challenges, intense political pressures and tight time constraints on the program, the panel commended the Pentagon and Boeing for having "formulated a sensible, phased, incremental approach to the development and deployment decision--while managing the risk."

But the panel also noted some management weaknesses. It said government program managers lacked clear authority, and it faulted Boeing for being too focused on the overall integration of the system at the expense of overseeing the development of the kill vehicle and other components. It urged Boeing to audit the ground testing of the kill vehicle more closely.

It characterized the kill vehicle program as "hardware poor," citing a lack of parts for both testing and backup. This hardware shortage, the panel said, had impeded development work and contributed to flight test delays.

For example, when the "inertial measurement unit" (IMU), which helps guide the kill vehicle, was found to be defective before last month's test, there were no spares. So the Raytheon Co., the kill vehicle maker, had to take an IMU from another kill vehicle designated to fly in a later test.

"Because the manufacturer has discontinued making IMUs, a new one cannot be substituted," the report said. "Rather, Raytheon must wait for repairs to be made," which could cause a flight test delay next spring. "This risk is directly attributable to inadequate developmental and spare hardware."

As important as flight tests are to the program, much of the data for assessing the system's feasibility by next summer will come from computer simulation and ground testing. Yet the panel found that these efforts had inadequate resources and were behind schedule.

"The plan to mitigate the risks associated with these delays did not provide much confidence to the panel," the report said.

The panel urged expanded simulation and ground testing. It also called for flight tests against more varied targets. And it recommended additional hardware to support testing, including a second launcher for the main test site in the Marshall Islands. Pentagon officials said a second launch silo is under construction.

Another "major concern," the panel said, is whether the kill vehicle will be able withstand the shock loads of the booster being designed to carry it into space. The expected loads would be more than an order of magnitude greater than those on the surrogate booster now being used in tests. Only in 2003 is the new booster due to be mated with the kill vehicle for the first time in a flight test.

CAPTION: Sen. Thad Cochran (R-Miss.) says, "We don't have the luxury of time. Because of the threat, we have no choice but to accept a high-risk program."