For more than two years, Pentagon officials have been kicking around the idea of sending one of their major acquisition programs through an extremely- lightweight version of the Defense Department’s usual oversight process in order to see what would happen if program managers were left to their own devices with a minimum level of red tape. Now, it appears as though they’re ready to take the plunge.
The notion of a “skunk works” approach to weapons system acquisition was first introduced in the 2.0 version of the department’s Better Buying Power program in 2013, but it hasn’t actually been tried yet because officials have struggled to find an acquisition program whose risk factors were sufficiently understood by its managers to make the Pentagon comfortable with a hands-off approach to oversight, said Alan Shaffer, DoD’s acting assistant secretary for research and engineering. He said DoD now has selected its first pilot project — a Navy program — though he declined to identify which one.
“To get to this kind of model you need to own the technical baseline, and we finally saw a program where the program manager owned the technical baseline,” he told the Potomac Institute on Friday. “If this works, it’s going to be the wave of the future. If we can cut out all the crap the program manager has to deal with to get a milestone decision and let them focus on actually managing the technical risk, we’re going to get better systems in a shorter amount of time. It’s not going to be a cookie-cutter solution: program management is hard, it’s complex and it’s dirty, but the bottom line is that you need someone who understands the program.”
While the skunk works method will comply with existing laws requiring review and approval by top Defense officials, it will push the Pentagon through those steps in a highly-streamlined fashion.
A long series of Pentagon meetings requiring months of preparation and documentation prior to each major program milestone will be replaced with a single on-site meeting in which the program’s entire status is reviewed and a decision is made about whether to proceed to the next step. The process, as envisioned, should take about three days.
“We’re trying to do something where we get the military service and the (Office of the Secretary of Defense) out of the program manager’s way and let them do their thing,” Shaffer said. “Right now, just for one Defense Acquisition Board, the military service whose program it is does three dry runs before their own service acquisition executive. And for each of those three dry runs, the program executive officers do a couple or three dry runs of their own. If you take a look at the poor program manager at the bottom of this thing, gravity works. They’re spending their entire day, every day, getting ready for higher-headquarters meetings instead of worrying about fielding the best system they can.”
Shaffer said DoD wants to see more widespread use of the model. And there may be more to come after the Navy pilot.
In the third iteration of Better Buying Power, released in March, Frank Kendall, the undersecretary of Defense for acquisition, logistics and technology ordered all three military departments to identify candidates within their own services to serve as skunk works pilots, citing the need for more rapid acquisition in order to staunch what the department feels is a continuing loss of U.S. technological superiority.
Hinting at the criteria DoD officials will use to decide whether programs are a good match for the streamlined approach, Shaffer said they are not necessarily looking for low-risk systems. Rather, they’re searching for instances in which the program managers have clearly shown they already understand the risks they’re dealing with, reducing the need for additional supervision and second-guessing from the rest of the DoD bureaucracy.
“In the case of the Navy program, the manager came in and said, ‘I don’t know if we’re going to be able to field this system successfully,’ Shaffer said. “It’s still in preliminary design and there are concerns about power and weight. But when the person talks about the technical risk, he understands it. He knows where the weight’s coming from, what weight could come out of the system if it works, he understands the interfaces and he has a very well-articulated developmental test plan that will tell us whether or not we’re going to make it. That’s what we need: understanding how a system works in exquisite detail.”
But reflecting on a 40-year Defense career during his discussion Friday, Shaffer lamented that program managers’ capacity to grasp the minutia of the systems they’re charged with fielding has waned through the years.
He prefaced his critiques by pointing out that the U.S. system still turns out unmatched technological capabilities, but said some of DoD’s past and high- profile problems, such as the failure of the Army Future Combat System and massive cost overruns in the Air Force’s ongoing Space Based Infrared System (SBIRS) program can be chalked up to a lack of technical know-how within government program management offices.
“We initially tried to field SBIRS with a political scientist as our program manager,” Shaffer said. “He was a great guy, but he had absolutely no clue when contractors were trying to blow falsehoods up his pant leg. It was just amazing. The program went out of control, and we wound up with several Nunn-McCurdy breaches. That’s why you have to own the technical baseline.”
He had a similar diagnosis of the Army’s FCS program, saying managers failed to understand the technical barriers to their grand vision, which aimed to link tanks, helicopters and unmanned vehicles into a seamless battlefield network, using, in part, wireless technology.
“It was a very complex system, and nobody in the Army sat down and realized that physics prevented the thing from working,” he said.
Consequently, Shaffer said he feels strongly that DoD program management offices need many more engineers at their helms.
“It pains me to say that as a mathematician,” he said. “But our program managers should probably be engineers. There’s a certain amount of rigor that’s needed to deliver a program, and the department got away from that. The Air Force used to have nothing but engineers [as program managers]. You have to own the technical baseline.”
To achieve that, Shaffer said DoD will need to continue to move to open systems architectures and own the interfaces between all of the major subsystems that make up a major weapons system so that new technologies can be plugged in, smaller suppliers can compete to offer their ideas, and the Pentagon can avoid locking itself into a single vendor in perpetuity.
He said DoD also needs more insight and control over the defense industry research and development it funds on a reimbursable basis, known as IRAD.
As part of 3.0 version of Better Buying Power, the Pentagon announced that companies would need to find a government sponsor before DoD would agree to pay for their research, something the large prime contractors have strongly resisted.
“Within the last three weeks, industry has been all over the congressional committees, telling them what dirty bastards the people in the Defense Department are for trying to influence their IRAD. I met with the big six contractors and told them that they can either work with us to come to a good solution, or they can keep going up the Hill and the staffers will keep telling me what they’re saying. That’s unproductive for all of us,” he said. “We’ve got to work with industry. They’ve got to make a profit, but we’ve also got to break apart the company-proprietary solutions. We need to go to openness and drop-in capabilities. We’re going to get resistance on that from the major companies, but at the end of the day, it’s going to be a tough market in the next couple of years. It’s going to be in their best interest to have as many competitions and competitive entry points as possible.”