By Geoff Fein
The Air Force’s effort to develop a satellite communications capability for airplanes and ground sites is proving to be a good example of what needs to be done to reform the acquisition process inside the Department of Defense (DoD), according to the Pentagon’s top acquisition official.
John Young, the Pentagon’s acquisition chief, told reporters last week that he sent a memo to Defense Secretary Robert Gates about the Air Force’s Family of Advanced Beyond line-of-sight Terminals (FAB-T) program.
“I am trying to keep him informed on decisions I think that need to be made,” Young said.
The gist of the memo, Young said, was that the services were all developing or using their own satellite communications terminals. The Navy was starting on the Navy Multiband Terminal (NMT), the Army was already ahead of the Navy and had a terminal called Secure, Mobile, Antijam, Reliable, Tactical Terminal (SMART-T), and the Air Force has FAB-T.
“You had a study done in the Navy that [asked] what are the chances for the services to do something together? And we find out that developing these satellite terminals costs about $1 billion,” he added.
The fact that all three services had their own unique satellite terminals was frustrating, Young said.
The Air Force undertook FAB-T in 2002, but the program had its beginnings back in the mid-1990s, Young noted. “The book was written on it from the mid ’90’s to 2002.”
“But when the book was written and the contract signed in 2002 the estimate was that that terminal, which was to be installed in airplanes and some ground [sites], was going to cost $270 million to develop,” he said. “Today the estimate is it is going to cost $1.6 billion to finish FAB-T. This is killing me.”
In 2006, it was recognized that FAB-T was going to cost closer to $1.4 billion. “There had been 150 contract changes on the contract,” Young said.
So the call was made for oversight from the Office of the Secretary of Defense (OSD), more discipline, a rigorous baseline and the need to execute the program, Young said. “It [has] actually executed reasonably well since ’06.”
He noted that if if wasn’t because of poor program foundation, no prototyping, and overly optimistic estimates, the program may not have seen such a huge cost increase.
“It might have cost $270 [million] if it had truly been a satellite communications terminal. But a lot of contract changes [occurred], a lot of changes about the interfaces with the airplanes, and the terminal itself changed to become a command and control device which actually can tell the satellite to change positions to support communications,” Young said. “It is almost like a small satellite ground station. That was never what was supposed to be done for $270 [million].”
Young questioned, however, whether that is truly cost growth.
“I potentially would say no. On the other hand, I would say the acquisition team should have stood up and said ‘hey, this isn’t want we agreed to do. We agreed to build you a radio, not a satellite command station,'” he said.
Young pointed out that DoD doesn’t have good management tools, and that is something that needs more work.
“When industry is finishing a program and finishing the work on a program, how can you make sure they take the people off the program? Frequently, our independent costers say you can only take people off programs so fast.”
That’s because industry working under “cost-type contracts” where the government is paying all of industry’s costs, does not rush to necessarily take people off programs, Young said.
“They do useful work, but…industry needs to put them on other jobs,” he added.
Some companies in the private sector will only put the number of people on a program they need to have on a program, Young said, so that they can efficiently and competitively compete in an aggressive marketplace.
“In DoD if I have a cost-type contract, you may or may not take the people off the program. So, it unfortunately becomes a requirement for us in a cost-type contract to budget for the likely reality of how fast industry will finish the work and take people off the program. That’s one of the things I explained to the Secretary of Defense in the memo.”
FAB-T could possibly finish for $1.4 billion or $1.5 billion, Young said. “But we don’t have a good way and we definitely don’t have enough people to go everyday [and] ask industry ‘do you need that person, and that person, and that person? Can you take them off the program now because you are finishing the work and you need to keep constantly leaning the program down, finish the test phase, and be done with the development.’ Those are the kind of things we struggle with.”
Acquisition reform will require DoD to buy systems at economic order quantity, stop changing the budget annually on multi-billion dollar development programs, and stop changing requirements.
“We’ve got, as an acquisition team, to build better initial cost and schedule estimates. We have to hold the line on requirements, which I am trying to help from the top…down. I need the program manager level and up…to do the same thing,” Young said.
“That’s why you hear Secretary Gates saying we don’t necessarily need another study. We know what needs to be done. We need to set reasonable requirements, use technology that’s adequately matured to be developed and manufactured to meet those requirements, build budgets that fully fund development of the program, not change budgets annually, and then do all the other blocking and tackling,” he said.
That would include certifying the technology readiness of programs, having good people in place to manage the programs, and use competition wherever possible, Young said. “Things like competitive prototyping to improve my knowledge. Competitive prototyping is as much…about putting that better foundation under a program.”
DoD needs to move away from programs starting at Milestone B based on thousand page proposals and view graphs that lead to guessing what the program is going to cost and how long it is going to take, Young said.
“If you want to know the heart of competitive prototyping, it’s about…[building] a piece of hardware so that at Milestone B we can base our estimates of future costs and our schedules and assess the maturity of the technology based on a piece of hardware that can better inform those estimates, so we can get out of this…’you made me guess the cost…I tried to go to Milestone B too early and then when I realized all the truths…the cost went higher,” he said.
“Because what I have said since I was in the Navy, the outcome of a contract has been known the day we sign the contract. The only questions–does my government program manager and his industry counterpart know that answer. For several years past we have not known those answers the day we have signed the contract, and I am trying hard to make sure we know those answers.”