4 November 2014

Follow These Steps to Reform the Way We Buy Weapons

Keep it simple, read a lot of books

The Internet loves listicles, those ubiquitous numbered lists masquerading as thoughtful articles on some topic that is essential to your health and happiness and just might change your life! I figure it’s about time somebody finally wrote a listicle about defense acquisition reform, which I’m pretty sure has never been done before.

Stop looking for one big fix

There are many sides to the problem of Pentagon overspending on delayed projects that are ill-suited to the military’s operational needs. There are lots of reasons weapon systems cost more, take longer and do less than promised.

The profound complexity of the problem means no single solution will once-and-for-all ensure the complete rainbow-unicornification of the military-industrial complex.

So let’s abandon the idea that today’s flavor of the month will succeed where all past attempts have failed.

The flip side of this means we should stop discounting and dismissing every reform approach that isn’t a cure-all. Let’s not criticize partial improvement concepts for leaving some portions of the problem unaddressed.

Once more time—there is no cure-all. If we’re lucky, we might find a cure-some. Even fractional improvement might be a step in the right direction.

Focus on delivering capabilities

In The Book of Five Rings, 16th-century ronin Miyamoto Musashi describes many postures, principles and techniques of swordsmanship, all of which he commends to the student with variations on the phrase “one must study hard.”

However, his most important piece of advice is as follows. “Even if one blocks, strikes, hits or touches the long sword of the opponent when he attacks, these are all opportunities for cutting the opponent … It is important to think of all things as a means for cutting.”

The acquisition community could learn something here. To paraphrase Musashi—“Even when one does paperwork, design reviews, funding drills or PowerPoint presentations, these are all opportunities for delivering operationally relevant capabilities.”

We can see hints of this wisdom in Under Secretary of Defense for Acquisition, Technology and Logistics Frank Kendall’s recent memo about the updated DoD 5000.02 instruction, when he warns against producing “compliance documentation” that is not actually useful “for planning and managing the program.”

His point is the documents we write should be helpful, not simply produced for the sake of passing an inspection.

Whether we’re talking swordplay or defense acquisitions, it is easy to get distracted by peripherals. One must study hard indeed to ensure our actions are focused on achieving the real goal.

Focus on the near term

Organizations like the Army’s Rapid Equipping Force have demonstrated the ability to quickly deliver technical solutions in response to today’s emerging threats. Since we can do that today, there’s a pretty good chance we can do it again tomorrow when the threat changes.

The key is to not only focus on building military gear for now, but to simultaneously foster an institutional capacity to consistently deliver relevant gear in a timely manner. In other words, the tactical ability to rapidly deliver relevant capabilities is itself a strategic capability and should be nurtured.

Prepare for a surprising future

The fact that the future will be surprising should not come as a surprise. Since nobody knows what’s around the corner, we need to build agility and flexibility into our systems rather than locking budgets and designs into 25-year programs.

Fortunately, there are ways to do this. Modularity, well-defined interfaces, and open architectures all help position a design to be responsive to unexpected developments—and these practices don’t cost more or take longer. In fact, when we look at total costs, they are often faster and cheaper.

Similarly, it is possible to establish realistically short schedules and tight budgets, making use of the actual amounts of time and money available, rather than relying on imaginary quantities of dollars and days in the distant future. Check out Federal Acquisition Record section 39.103—on modular contracting—for some specifics.

At top—then-Defense Secretary Robert Gates tours the Texas plant that makes F-35 fighters in 2009. Above—U.S. Army armored vehicles in storage. Defense Department photos

Stop using the waterfall method

Repeat after me—the sequential design approach to software development, known as the “waterfall method,” does not work. It never did. Never ever, ever.

In fact, when Winston Royce formally described this approach in 1970, he was using it as an example of how not to do things. Royce’s paper explained the approach is “risky and invites failure” and concludes that it “has never worked on large software development efforts.”

Why so many people immediately adopted it as a preferred practice is a mystery for the ages. Probably they only read page two of Royce’s paper, where he describes the method, and didn’t get to page three or beyond, where he says it’s the wrong way to do it.

Need more convincing? How about a 2000 Defense Science Board report which pointed out “About 90 percent of the time, the [waterfall] process results in a late, over-budget, fragile and expensive-to-maintain software system.” Pretty sure that’s exactly what Royce said more than 40 years earlier.

Instead of using the method which has been known to not work for at least 43 years, the Pentagon could get serious about Agile. Or Scrum. Or Extreme Programming. Or anything else at all. There is simply no excuse for continuing to climb into a barrel and float over Niagara Falls.

As John Wayne told me last week while we sipped whiskey around a campfire, “Software is hard. It’s even harder when you’re stupid.” Please, let’s not be stupid. And yes, using the waterfall method on large software efforts is, to use the technical term, stupid.

Read the FAR—seriously, read it

Allow me to quote section 18.101 from the Federal Acquisition Regulation. “The FAR includes many acquisition flexibilities … These acquisition flexibilities do not require an emergency declaration or designation of contingency operation.”

So the next time someone tries to tell you the regulations absolutely, positively can only be interpreted in a single, narrow, confining manner or tries to argue that an alternative approach requires extraordinary circumstances, have them read section 18.101, then politely ask them to get the heck out of your way.

Alternatively, we might want to check out section 13.003—“Agencies shall use simplified acquisition procedures to the maximum extent practicable.” Or any other section that comes to mind—there are plenty of good ones to choose from. And thanks to the Internet, they’re not hard to find.

The key is to make the policy work for us, which isn’t hard to do once we actually look at the policy itself.

The policy may be stupid, but being stupid is not the policy

Speaking of policy … In the occasional situation where the FAR is less than friendly or seems to require behaviors that run counter to common sense, it’s okay to apply a little intellect.

More than okay, in fact. It might even be important or essential to do so. Oftentimes all that’s required to avoid launching onto a course of action that everyone agrees is dumb is a little creative problem solving and a little thoughtful interpretation. Even the most inflexible-sounding bit of guidance can often be executed or implemented in a variety of ways.

Process is neither the problem nor the solution

The much maligned defense acquisition is only a symptom of a deeper dysfunction. Replacing one process with another is not going to matter if we continue to interpret the new process through the same dysfunctional lens.

As long as we treat compliance as a higher virtue than capability, view complexity as a sign of sophistication, consider budget overruns as inevitable, and use schedule delays as a desirable problem solving method, it really doesn’t matter what the process says. We’re going to take longer, spend more and deliver less than promised.

So instead of focusing on the definition, documentation, or design of the process, let’s make sure our attention is persistently on the actual goal of delivering affordable systems that are available when needed and effective when used. See earlier comments about Musashi.

Read a book

Might I recommend starting with Jim Burton’s The Pentagon Wars? No, watching the movie doesn’t count, although it’s better than nothing. Then check out Robert Coram’s Boyd. Musashi’s Five Rings isn’t a bad choice either.

Other must-reads include Garr Reynolds’ Presentation Zen, Clayton Christensen’s The Innovator’s Solution, Howard McCurdy’s Faster, Better, Cheaper, Gordon MacKenzie’s Orbiting the Giant Hairball, Tom Peters’Reimagine, and Don Norman’s The Design of Everyday Things. Also, Skunk Works by Ben Rich. I almost forgot Alex Laufer’s Project Management Success Stories. And Octave Chanute’s Progress in Flying Machines. That’s a good start.

My point is there is a lot of really important stuff out there, stuff that will help equip us to make good decisions. Make sure you read some of it. And don’t miss Henry Petroski’s To Engineer Is Human. Alright, I’m really done now.

Dan Ward is a lieutenant colonel in the U.S. Air Force, currently stationed at Hanscom Air Force Base. He is the author of FIRE: How Fast, Inexpensive, Restrained and Elegant Methods Ignite Innovation, published by HarperBusiness. The views expressed in this article are solely those of the author and do not reflect the official policy or position of the Air Force or Department of Defense.

No comments: