F-35 Production Costs, Line Changes Drop: Lockheed

Fort Worth: Lockheed Martin’s mile-long aircraft factory here sent the the twelfth F-35 Joint Strike Fighter produced this year to Eglin Air Force Base last Wednesday. Though no cause for champagne, the delivery marks an important milestone in the company’s efforts to ramp up production. The plane took less than half as many touch-labor hours to assemble as did the first two Air Force versions, both of which came off the production line on May 17 last year.

Company officials, about to enter negotiations for the fifth tranche of low rate initial production (LRIP 5), say the dramatic reduction over the past 18 months in the amount of time it takes Lockheed mechanics to put together one of the complex stealth fighters is a sign the program is recovering from the crisis it faced in February 2010. That’s when former Defense Secretary Robert Gates fired the government’s F-35 program manager and withheld $614 million in payments to Lockheed because of cost overruns and schedule delays in the project, whose goal is to build more than 2,000 fighters for the United States and 10 other countries.

“We’re seeing learning curves that are really the best that have been seen in this industry at this point,” says Larry Lawson, executive vice president and general manager of the F-35 program for Lockheed. “We’re getting incredible performance in reducing those hours.”

Lawson ran Lockheed’s F-22 Raptor fighter program from 2004 until the company moved him to the Joint Strike Fighter in August 2010. When I interviewed him in his Fort Worth office Oct. 19, he hehehhehehhhjkkllwas beaming over the good results of sea trial flight tests of the F-35B, as the STOVL (short take off/vertical landing) variant being built for the Marine Corps is designated. Those trials, which were still going on aboard the amphibious assault ship USS Wasp at the time, ended Oct. 21 after two F-35Bs successfully completed 72 vertical landings on the ship.

Lawson emphasized that it’s up to the government to decide how and when the F-35B can get off the undefined “probation” Gates declared it to be on last January because of engine, weight and potential cost problems, but the executive said he was confident Lockheed had solutions in place or in process that would resolve those issues.

“Corrective action” has been taken, he said, to solve problems with an innovative “lift fan” behind the F-35B’s cockpit. The fan’s 50-inch diameter counter-rotating blades turn when engaged by a clutch connected by a drive shaft to the jet’s Pratt & Whitney engine, creating a column of air that produces 18,000 lbs. of thrust under the plane as it hovers or lands vertically. A swiveling engine exhaust nozzle at the aft of the F-35B directs another 18,000 lbs. of jet thrust downward while two “roll post” nozzles in the wings each funnel down yet another 2,000 lbs. of vertical thrust to provide lateral balance.

Lockheed has also redesigned and next year will flight-test two doors that open behind the lift fan atop the fuselage to provide extra air for the engine when the B variant is hovering, Lawson said. The original auxiliary air intake doors — still being flown on F-35Bs performing flight tests — oscillate when open, creating no safety problem but making it likely they will wear out far earlier than they should.

The company has also gathered “a lot of data,” Lawson said, that will enable engineers to come up with ways to mitigate a phenomenon called “suck down,” in which turbulence under the F-35B as it lands vertically can create a vacuum that pulls the plane down too rapidly — a potential danger, especially for a pilot returning from a mission with unexpended bombs or missiles.

The final assembly learning curve (as the reduction in the touch-labor hours it takes to put the aircraft together is called) has come down rapidly and should keep falling as Lockheed builds the next few lots of F-35s because the factory floor has finally absorbed production changes required when the B variant was redesigned five years ago, Lawson said. The redesign was required to shed the STOVL plane of 3,500 lbs. of unacceptable weight.

“When you have a high amount of change in the system, that churn creates problems in your supply base, which creates shortages to the factory,” Lawson explained. One result of such supply shortages was that certain steps in assembly had to be done out of order, meaning they were being done at the wrong work station, on the wrong industrial tool and often by the wrong workers. That, obviously, slowed the assembly line down. Now, however, the F-35 line is not doing such “out of station work,” he said.

Don Kinard, deputy director of the F-35 Fighter Production System — i.e., the factory floor — said the latest F-35A took about 110,000 touch-labor hours to assemble rather than the roughly 250,000 hours the first A variants required.

One reason for this steeply falling learning curve, Kinard said, is that Lockheed’s F-35 mechanics are simply learning to put the planes together faster. That’s no surprise. Just as consumers who buy build-it-yourself furniture find it easier to put a bookshelf or table together if they’ve done one before, aircraft assemblers — like other industrial workers — need less time to do their tasks as they repeat them, up to a point.

Other reasons for the learning curve decline include the fact that a large amount of the F-35 assembly work is automated — done by robotic machinery. Finally, the company constantly works to improve assembly processes.

Along with an increase in the number of aircraft ordered in Low Rate Intitial Production Lot 4 as compared to LRIP Lot 1, the learning curve drop is one reason the plane that left Fort Worth on Wednesday cost $111 million, roughly half the price the government paid for each of the first two LRIP F-35As.

“What to me is remarkable is that with three different variants, we’re still able to maintain a legacy learning curve,” Kinard said, meaning a curve that is falling at roughly the same rate as did the learning curves for Lockheed’s single-variant F-22 and F-16 fighter planes. While 100 percent of mission systems, such as avionics and software, are the same in each variant of the F-35, the conventional takeoff A variant, the STOVL B variant and the C variant for use on aircraft carriers share only 20 percent common structure, Kinard said. The F-35C has a larger wing and heavier landing gear than the other two variants, for example, while the B variant is the only one with the lift fan and boasts a smaller weapons bay than the other two.

At the moment, Kinard said, it takes the factory two years to assemble an F-35. In the future, the learning curve should bring the amount of touch-labor hours down to about 50,000 and reduce the span of time the factory needs to produce a single aircraft to somewhere between 12 and 14 months.

Whether Lockheed will need to build them that fast anytime soon, however, is uncertain. The day after the latest F-35A was flown to Eglin Air Force Base, the government program office announced that, to cover previous cost overruns, five aircraft — four for the Air Force and one for the Navy — will be cut from the next and fifth production lot, making it 30 instead of 35 aircraft expected to be bought this year and built by 2013.