By Renae Merle
Washington Post Staff Writer
Wednesday, February 7, 2007
After 10 years and $1.7 billion, this is what the Marines Corps got for its investment in a new amphibious vehicle: A craft that breaks down about an average of once every 4 1/2 hours, leaks and sometimes veers off course.
And for that, the contractor, General Dynamics of Falls Church, received $80 million in bonuses.
The amphibious vehicle, which can be launched from a ship and then driven on land, is so unreliable that the Pentagon is ditching plans to begin building the first of more than 1,000 and wants to start over with seven new prototypes, which will take nearly two years to deliver, at a cost of $22 million each.
The Expeditionary Fighting Vehicle is one of the Pentagon's largest weapons programs and exemplifies the agency's struggle to afford a cadre of new mega-systems that are larger and more complex, but also more trouble, than their predecessors.
Despite reforms meant to rein in costs, it is not unusual for weapons programs to go 20 to 50 percent over budget, the Government Accountability Office recently found. Among the offenders is the Army's sprawling modernization program, which aims to update everything from tanks to drones and is now expected to cost $160 billion, up from $90 billion, and a Lockheed Martin missile-warning satellite program, which is projected to cost more than $10 billion, up from $4 billion.
The Marines' troubled program is on a collision course with critics who are wary of its growing price tag and who wonder about the utility of an amphibious vehicle meant to storm beaches in a way the military hasn't done for decades, at a time when soldiers are consumed with urban warfare in Iraq and Afghanistan.
The Marines, though, have not been dissuaded by the vehicle's poor showing in tests, as such fits and starts are common in military development programs.
"We were disappointed. We weren't shocked," program manager Col. John Bryant said.
The cost of the amphibious vehicle effort has increased 50 percent, to about $12 billion from $8 billion, with another cost bump projected after the program is relaunched.
The overruns are eating away at the Pentagon's buying power but not its appetite. The amount the Pentagon plans to spend on major weapons systems has doubled in the past five years, to $1.4 trillion from $700 billion, according to the GAO.
"I would never state, in an enterprise this large, that we ever have it all under control," Kenneth Krieg, the Pentagon's acquisition chief, said in a December interview. "I think we're on a good path and only performance will prove that, and that's what we got to do."
When it was launched in 1996, the Expeditionary Fighting Vehicle was promoted as an example of acquisition reform as hundreds of General Dynamics and Marine Corps officials moved into the same 62,000-square-foot office building in Woodbridge to run the program, in hopes of saving time and money. The program's efforts to keep maintenance costs low won plaudits from Defense leaders and twice earned the program the Pentagon's highest acquisition award, in 1997 and 1999. In 2001, the program collected an innovation award for developing a system to keep the craft's internal components from overheating, a technology that has been adopted for other weapons.
But the program has struggled with repeated delays, cost increases, budget cuts and dashed expectations, according to military officials and government reports. Problems range from leaks in hydraulics systems to software glitches, according to the reports. Last year, the vehicles completed just two of 14 planned tests.
"They started out really well, and I was really pleased," said Philip Coyle, the Defense Department's former director of operational test and evaluation. "But gradually the complexity of the program has overcome the contractor, so they are years behind schedule."
General Dynamics defends its progress, noting that the vehicle has met many goals, including being able to reach speeds of 30 knots on the water. The vehicle is fast enough to keep up with the Abrams tank on land, it can carry 17 Marines, and its systems can communicate with other ships and tanks, all key performance criteria, the company says.
"I have heard no one in leadership say that they don't need this capability, that they don't want this capability," said Peter Keating, a General Dynamics spokesman.
An independent review released in December by the Navy's acquisition office questioned the company's commitment to solving the development problems that plagued the vehicle. The report said General Dynamics appeared more interested in starting production than trouble-shooting and didn't manage the groups making many of the decisions. The production phase is typically more profitable for a contractor and often marks a point at which a program becomes more difficult to cancel.
General Dynamics "seems to be focused on production rather than on solving significant design and engineering problems," the Navy report said. "This must be changed if the Program is to move ahead successfully."
Noting that the bonuses the company received did not reflect the vehicle's performance, the report recommends that the Defense Contract Management Agency consider recovering some of the award fees. A DCMA spokesman said the agency had not addressed the issue because it was unaware of the recommendation.
The Marine Corps has tried to deal with some of its development problems by lowering its expectations for the vehicle. The service originally wanted a craft that could operate 70 hours between major breakdowns, but it cut that target to 43.5 hours after tests revealed the vehicles were struggling to meet the higher goal.
But even the lower targets have been hard to hit. Marine Corps officials were distressed to discover that the prototypes encountered an "operational mission failure" on average every 4.5 hours in tests last year. There were 645 failures within the subsystems, overwhelming the three-man maintenance crew, according to a report by the service's testing agency.
"It's a very complex vehicle; when it breaks, it's difficult to repair," said Col. Michael Bohn, director of the Marine Corps testing agency.
General Dynamics has also struggled with the vessel's complicated software. During testing, the software operating the guns didn't always fire on command, for example. Some of the problems date to 2001, when responsibility for software development was given to a General Dynamics division with little track record of running such a complex project, according to the GAO. The Marine Corps gave the company a year to improve the division's capabilities, the report said. When it failed to meet the deadline, the company was granted an extension. Eventually, the Marine Corps turned 70 percent of the work to a government organization.
"From a taxpayer standpoint, I think, 'Gee, didn't we pay GD to do this?,' " said Paul L. Francis, the GAO's director of acquisition and sourcing management. "Are they going to pay us for doing it for them?"
Keating, the General Dynamics spokesman, said the company is confident it can meet the Marine Corps' demands. He noted that the service had asked the firm to establish a new software-development organization and that such organizations often take several years to mature. "We started from ground zero," he said.
Now, the Marine Corps is ready to start over. The service is to present a plan to salvage the program to Defense leaders in March. In addition to buying seven new prototypes, the Marine Corps proposal will probably require at least a two- or three-year delay, adding $200 million in development costs a year, said Bryant, the Marines program manager.
The service will seek to simplify the design, "getting rid of the complexity where we don't need complexity," he said.
Noting that his son was a Navy combat corpsman in a battalion using the aging amphibious vehicles that were supposed to be replaced by now, Bryant added: "I need to fix our reliability and make the EFV a vehicle I'd be proud to hand to my own son."