Skip to Main Content
Project on Government Oversight


POGO in the News

Power Trips, Part 2: Who's paying
June 5, 2006 

Battle on Border Turns Deadly
May 18, 2006 

Pilotless Spy Plane Plagued by Flaws
February 2, 2002 

More »



Fighting with Failures Series: Case Studies of How the Pentagon Buys Weapons - Predator UnManned Aerial Vehicle

March 25, 2002 


The Predator unmanned medium altitude "long dwell" aerial vehicle is operated by a ground pilot positioned in a van as far as 400 miles away -- flown primarily in areas of "moderate risk" from being shot down by enemy fire. The aircraft, equipped with satellite-linked "real-time" video cameras and sensors, is primarily used for surveillance, reconnaissance, and target acquisition. However, media reports indicate that in recent months the Central Intelligence Agency has fired Hellfire guided missiles at least twice from Predators wings during combat missions in Afghanistan, although it was never designed or adequately tested for this use. The Air Force maintains that such use of the aircraft as an offensive weapon is only experimental. Despite glowing reports about the Predator's overall effectiveness in Afghanistan, the Pentagon has offered no data to back up such claims.

The Predator was developed as a special demonstration project in 1994. The aircraft was evaluated by the Director of Operational Test and Evaluation, who declared it "not operational effective or suitable" in September 2001. The Air Force originally ordered a fleet of 48 Predators. Each system of four aircraft and a mobile control center requires a support team of 55 Air Force personnel and two civilian technicians to operate and maintain.

Total number of systems: 12 (48 aircraft)
Total Program Cost: $640.9 Million
Average Unit Cost: $25 Million (van + 4 aircraft)
Contractor: General Atomics Aeronautical Systems, Inc., San Diego, California


  • Although Pentagon officials say the Predator has been a very valuable asset during the war in Afghanistan, so far, it has proven to be a fair-weather aircraft. It cannot be launched in adverse weather, including any visible moisture such as rain, snow, ice, frost or fog; nor can it takeoff or land in crosswinds of greater than 17 knots. Although each Predator system has two sets of "weeping wings" de-icing devices, so far, the wings have not yet demonstrated they work well, as underscored by the November 2, 2001 crash of a Predator in Afghanistan due to icing.

  • While it was deployed in the Bosnian theater in 1996-1997, nearly half (226 out of 479) sorties were canceled or aborted because of weather.

  • Predator experienced similar problems during Operation Allied Force in Kosovo, where weather conditions permitted unimpeded air strikes for only 24 days of the 78-day air campaign in 1999.

  • Investigators have attributed more than half of the 25 Predators that have crashed to mechanical failure, weather, or operator error. Although more than 40 percent of the Air Force's original Predator fleet has been lost, military officials say such losses are justified because it does not expose pilots to risk of injury, enemy capture, or death.

  • Because it cannot evade radar detection, flies slow, is noisy, and must often hover at relatively low altitudes, the Predator is vulnerable to being shot down by enemy fire. In fact, an estimated 11 of the 25 Predators destroyed in crashes reportedly were caused by enemy ground fire or missiles.

  • The Predator has failed to meet requirements and has not proved to be an effective targeting tool. "Because of deficiencies in communications capability and target location accuracy, the Predator did not show utility for missions such as strike support, combat search and rescue, or area search," the Pentagon's chief tester concluded in his September 2001 test report of the aircraft.

  • Although it carries a camera for night surveillance, testing at 30,000 feet showed that the Predator could only recognize potential targets during daylight hours and clear weather. During testing, even during daylight hours, the aircraft's cameras could not, for instance, distinguish tracked vehicles such as tanks from wheeled vehicles from higher elevations. "The Predator's infrared camera could detect targets, but could classify [between wheeled versus tracked targets] only 21 percent of the time and recognize [the model of tank, i.e., Russian versus American] only 5 percent of the time. The Day TV camera could detect targets, but could not classify or recognize them," said the DOT&E report on the Predator.

  • The Predator has very limited value for classifying or recognizing targets due to its poor quality of camera resolution. To provide any level of detail, it must switch to its spotter camera, which has such a narrow field of view that the aircraft loses its situational awareness."Looking through the Predator's camera is somewhat like looking through a soda straw," Philip Coyle, the Pentagon's former chief independent tester, told Slate Online Magazine.

  • Because of all these deficiencies and reliability problems, the Pentagon's chief tester, Thomas Christie, Director, Operational Test and Evaluation, concluded that the Predator is not performing as promised, declaring that the aircraft is ineffective and unsuitable for military operations. Air Force officials say the test data is old and that they have corrected some of the deficiencies identified in operational testing, but have declined to say which ones. "They always say that," Coyle said.

Short Cuts on the Way to Acquisition:

  • The Predator's lightning speed engineering and development has been a double-edged sword. Because it progressed from a concept to deployment in less than 30 months, the drone was sent into a combat zone before it was tested for battle readiness.

  • Although Predator flights began in 1994, full operational testing -- a series of realistic battle tests designed to weed out bad weapons -- of the Predator slipped behind schedule by more than three years, not commencing until October 2000. If operational testing had been completed before production, many initial glitches and costs increases might have been averted.

  • The Director of Operational Test and Evaluation reported that the Air Force did not take full advantage of operational assessments made from 1995-2000 that identified key system shortfalls. In fact, many were repeated during initial operational testing in late 2000.

  • After it completed testing in the developmental stages, the Air Force Operational Test and Evaluation Center concluded that the Predator system did not meet reliability or maintainability requirements and displayed a "high preventative maintenance burden." Still, the test center certified the Predator as battle worthy.

POGO's Fighting with Failures Series documents Pentagon shortcuts in testing and operational requirements that have resulted in weapons that do not work, that waste taxpayer dollars, or that are not suitable for combat.

Sources: Fiscal Year 2001 Annual Report, Director, Operational Test and Evaluation, February, 2002; Operational Test and Evaluation Report on the Predator Medium-Altitude Endurance Unmanned Aerial Vehicle, Director, Operational Test and Evaluation, September 2001; Fiscal Year 2000 Annual Report, Director, Operational Test and Evaluation, March 2001; NATO'S Air War For Kosovo: A Strategic and Operational Assessment, Benjamin S. Lambeth, RAND, 2001; Dull Drone: Why unmanned U.S. aerial vehicles are a hazard to Afghan civilians, Eric Umansky Slate Online Magazine, March 13, 2002. 

Founded in 1981, the Project On Government Oversight (POGO) is a nonpartisan independent watchdog that champions good government reforms. POGO's investigations into corruption, misconduct, and conflicts of interest achieve a more effective, accountable, open, and ethical federal government.

# # #