The well-known valley of death between the DARPA or Military Service Science & Technology (S&T) development and military Programs of Record (PORs) is the result of the high-entropy, high-noise channel, that sits between S&T and the bureaucratic DoD acquisition system. This noisy acquisition demise can be traced backward from today’s Planning Programming Budgeting System (PPBS), Federal Acquisition Regulations (FAR), Operation of the Defense Acquisition System Instruction (DoD Instruction 5000.2), and Joint Capabilities Integration and Development System (JCIDS). Each are now large bureaucracies that help make up the majority of the 25,000-person army of centralized DoD oversight, operating from the Pentagon.
The root of these centralized policy and processes changes can be traced back to Secretary of Defense Robert McNamara‘s Whiz Kids, introduced into the Pentagon in 1961. Using his automotive executive background he brought modern economic analysis, operations research, game theory, computing, and modern management systems to the Department. It was his invention of the PPBS that was supposed to introduce unprecedented budget transparency and pinpoint management responsibilities for weapon system acquisition.
Since the introduction of McNamara’s PPBS, each of the additional high-noise policy and process changes have further centralized Defense acquisition in the name of greater efficiency and tax dollar effectiveness. The most sweeping changes came from the 1986 Goldwater-Nichols (GN), Department of Defense Reorganization Act which passed into law two major changes. First it decentralized military operations by streamlining the chain of command from the President through the Secretary of Defense directly to Combatant Commanders, thereby bypassing the Service Chiefs. Second it centralized acquisition by directing the establishment of the Office of the Under Secretary of Defense for Acquisition, USD(A), similar Component Acquisition Executives (CAE), and Program Element Officers (PEOs) that manage groups of related Program Managers. This sweeping acquisition change finalized the full separation of acquisition from the military Service Chiefs, who were left with the authority only to set requirements and provide acquisition budgets. This GN acquisition-side change has led to the unintended consequences of today’s high-entropy, high-noise, over regulated acquisition system that needlessly squanders National Security budgets for long timeline, too-big-to-fail acquisition programs such as the most recent F-35 Joint Strike Fighter.
Why do we know that Gilder’s bureaucratic noise theory reduces acquisition effectiveness? One only has to observe that every Secretary of Defense since Goldwater-Nichols has announced their intention to improve acquisition inefficiencies.
“Today’s defense acquisition system is a product of decades of reform initiatives, legislation, reports and government commissions. Major reform efforts began in earnest in the 1960s with Secretary of Defense Robert McNamara. His main reform efforts centralized control within the Office of the Secretary of Defense (OSD) and created the Planning, Programming and Budgeting System for resource allocation. Throughout the latter half of the 20th century, each administration left its own mark on defense acquisition, focusing primarily on the acquisition process itself, as well as Department of Defense management. However, many of the reforms recycled various schemes to shift decision-making authority from the services to OSD, realign oversight and accountability responsibilities, and alter the process (adding and removing milestones, phases and so forth). Despite these initiatives, cost and schedule growth continue.”The New Acquisition Reform Effort: Back to the Future – William Lucyshyn
The result has been several rewrites of the DoD 5000 instruction and the system hasn’t gotten better… but it has gotten worse! In addition, the Department has better trained the acquisition workforce through the Defense System Management College (DSMC), now Defense Acquisition University (DAU), where all PMs and PM staff members are required to have completed DAU training. Despite all of these well-intentioned improvements, the acquisition oversight staffs have gotten bigger, defense contractors have grown larger staffs to address the required oversight documentation, and an entirely new form of support-contractor has emerged to support program offices, in order to feed the oversight documents and process requirements. And after all of that, defense system acquisition can show no measured improvement in delivered and effective systems. But it can show performance, cost, and schedule failure after failure!
As discussed in Part 1 of this post, we can understand from DoD’s history, that a previously lower-noise, lower-bureaucracy acquisition channel enabled the human inspired entrepreneurial surprise of nuclear powered SLBM submarines developed and delivered in less than a decade, or AEGIS multi-warfare missile defense ships developed and delivered in thirteen years, just to name two pre-GN programs. Post GN, the only remarkable acquisition achievements have been made as special programs like the F-117 stealth aircraft that achieved IOC in six years following the DARPA proof of concept HAVE Blue aircraft. This program avoided the standard high-noise acquisition path by leveraging DARPA’s low noise development activities and extended that through special low-noise, high-profile, acquisition activities.
If there is any good news for future DoD acquisition it is that Congress, through the National Defense Authorization Acts (NDAAs) from 2016-2018, have passed several new laws designed to regain DoD’s technical edge and empower Program Managers to speed up acquisition and take advantage of as much commercial technology development as possible. The result of these law changes has been the breakup of the Under Secretary of Defense for Acquisition, Technology & Logistics, into two Acquisition Under Secretaries, one for Research and Engineering, and the second for Acquisition and Sustainment.
The goal of the USD Research and Engineering, the Department’s Chief Technology Officer, is to empower the adoption of breakthrough technologies, extend the capabilities of current war fighting systems, counter strategic surprise, and develop policies for rapid technology transition. One of the most empowering changes is NDAA Section 804 that authorizes Middle-Tier acquisitions and the use of Other Transaction Authority Agreements for rapid prototype developments. Middle-Tier acquisition does create a low-noise channel and allows for prototype to IOC development in five years using OTA Agreements and if successful authorizes full production under OTA for IOC to Full Operational Capability (FOC) in five years under an OTA Agreement.
The more traditional role of USD Acquisition and Sustainment supervises all DoD acquisitions including procurement of goods and services, while empowering Military Services by moving all major Service programs under acquisition control of the Component Acquisition Executives. This change reduces the drive for Joint programs and reduces some of the bureaucratic noise by leaving the Services in control of their high value acquisition programs. What is not addressed by these changes to date, is that all acquisition remains under control of the CAE and continues the GN practice of no Military Service Chief acquisition involvement with the materiel their forces must use to fight and win.
While these recent NDAA law changes are working to lower the acquisition process noise, they could further reduce the noise channel by putting the PEO’s and PM’s back under system commands or a single senior military acquisition head reporting to both the Service Chief’s and the CAE. Such a change would empower military leaders to get more engaged in acquisition activities while giving them the new tools of Middle-Tier acquisition to push major program acquisition back toward the post WWII timelines.
In the meantime, Military Service leaders are challenged by the potential for Artificial Intelligence (AI) to become a new battlefield game changer. While they cannot yet directly influence the acquisition of AI capabilities, they can help to align requirements and funding support toward Middle-Tier AI acquisitions. Given the near-peer pressure on the military adoption of AI, this is an area that needs a low-noise acquisition channel if it is going to effectively help introduce AI capabilities into military platforms, command & control systems, and weapon control systems.
As I have written in a previous blog post, Navy Acquisition Can Easily be Fixed, lowering acquisition bureaucratic noise could begin with these easy to implement policy/practice changes that require no Congressional approval:
- Speed to Capability;
- Passive Oversight;
- OTA Prototype to Production; and,
- Strong Technical Leadership.
These low-noise channel changes would reduce oversight, help deliver stable funding for shorter development timelines, empower high-performing program managers with full authority and accountability, while reducing the inefficiency of drawn-out multi-year contract awards, followed by the inevitable FAR contract protests.
… and because commercial AI is making such significant progress, AI is an excellent place to apply low-noise acquisition practices to ensure timely sustainment of our Nation’s military capability.
Thanks for the excellent article. I certainly hope that current procedures, and technical support can be used to remove all the problems that have bee ours for a long time. While I hope that all will be perfect, experience has pointed out to me that we will fix all the old problems, and in the process discover, implement, and become irritated with the new problems.
Service labs and PEO’s constrain innovation and speed. They use DOD process and PBSS to disrupt creativity. An example is DCGS. Fielded in 1999 Navy has spent billions and still doesn’t have capability at sea. Shameful performance.
Good point Phil, I remember well your work in DCGS-N and I am still working with them today…
Hi Marv, very insightful.
Many issues encountered with very large DOD programs is how they are led and managed from the start. When the Govt to Prime contractor relationship is not collaborative and transparent in dealing with the many technical and political challenges that will come up, oversight tends to catch issues late in development cycle. The result is replanning, rework, disruption, late changes, and overruns.
In late 80’s, a few large DOD programs adopted the Boeing 777 design/build, improve product and process development IPPD approach with success. This process encourages early involvement of all stakeholders to formulate a producible, supportable solution that meets achievable requirements in a collaborative manner. Just pointing out how a program is run can be a major contributor to its success.
Thanks Tim, good comments about how we run programs… my concerns are more about the excessive oversight by generally unqualified staffers that keeps the programs too busy to work the day to day program operations with the prime as you are suggesting.
An outstanding article, Marv! My comments address a few salient implications for further investigation in your blog.
1. Some of the least visible but most impactive programs for rapid operational prototyping and deployment have been the U-2 aircraft with its ground stations, the TACP CAS software upgrades and quiet C2 initiatives of many kinds. These efforts have permitted on-ramp and off-ramp of a wide variety of capabilities the warfighter has desperately needed over the years.
2. Offshore developments supporting gap-filling through later imports have successfully shunted our acquisition processes successfully, but they are tricky. Israeli programs for missile defense, advanced cruise missile defense, geospatial imagery and artificial intelligence, to name a few, continue to be a two-way exchange with value added at each transfer.
3. Breakthrough technologies for our major defense contractors can be fostered through massive—and fully funded—offshore developments with implications for many programs that lack momentum or sponsorship, or both. Such programs include securing oil and gas delivery systems, improved electronic border surveillance and control systems and space “junk” removal capabilities. Here our major defense contractors have to step out of their traditional swim lanes. A significant impediment to progress is industry’s reluctance to push out the tech envelope!
4. The myriad covert developments, to which your post chiefly applies, have the benefit of classification and caveat protection. Yet transitioning from black to white developments can be nearly impossible. Here we are also trammeled by a whole new range of markings for unclassified information.
Stlll, intelligence and energy can overcome all obstacles. Let’s just hope desperately needed capabilities arrive in plenty of time for ‘the next war’ as the international situation suggests we have no time to lose. After all, we will fight a ‘come as you are’ battle with a 360-degree threat in any case. Talent and long experience are waiting for the call to arms. We are fortunate many graybeards are still alive and active while our policy wonks flail and fail, the fit though few are already stepping forward in the shadows.
Thanks Wick… unfortunate that our success have to come from non main stream activities rather than be the normal… but you are spot on…
So, let’s make the non-mainstream the norm. Or take what works from the non-mainstream.
Exactly what I am proposing, at least for the non-platform systems… but even successful aircraft like F-117 used the DARPA built prototype to move quickly to the first IOC aircraft… Thanks Linda
Good article. Another process/vehicle for moving quickly is the Small Business Innovation Research (SBIR) program. This program provides for a highly competitive environment upfront and streamlined acquisition downstream. It has provided some successful key USN disruptions such as the Acoustic – Rapid COTS Insertion (ARCI) program; which revolutionized how cutting edge submarine sonar capability is acquired and developed.
good point Eric and I agree… but when it comes to spending big dollars we need to make sure our Government spending efficiency is much higher than the over regulated acquisition environment we see today…
“…nuclear powered SLBM submarines developed and delivered in less than a decade, …”
The boomers needed navigation services. In this case, Navy turned to USCG for the job.
First, how not to do it. The mine warfare community has, several times, attempted to build it’s own precision navigation systems. All went into black holes (eventually the argument was eclipsed by GPS and differential GPS). But that’s ahead of the story.
After WWII, we had an infrastructure of several Loran chains (today referred to as Loran A) which was 2nd generation technology (adapted from the Brits’ GEE system). That infrastructure remained fairly static for a decade. There were good technical innovations that could be made but the impetus to do so lacked until the FBM submarine program catalyzed. These included sampling the signal at the 30usec crossover point rather than at signal start — a the highest power point in the pulse curve. And settling on 100kHz as the frequency and differentiating chains by pulse group repetition frequencies. Moving from MF to LF allowed a fewer number of longer range stations for equal coverage. Also allows reception (with a towed wire antenna) by a submerged submarine. Shift from rubidium clocks and manual control to cesium clocks and an automated system area monitor system control (removes a good deal of opportunity for human error. But all these remained in the lab until …
… Navy tasked USCG to build five chains, the world’s first Loran C deployments, and supported the necessary (USCG) budget in Congress. Four of the chains (NW Pacific, Bering Sea, Med Sea, North Atlantic) were all directly covering the Polaris operating areas (the fifth was over Hawai’i and the islands are in the wrong places — the chain geometry yields poor coverage — it was billed as a ‘training chain’).
I’m unaware of the internal programmatic details, but the chains were all available by the time the first boomers sailed.
The side effects were, er, interesting. There are two major parts of this saga:
1. The first is that as soon as you turn on a chain, ship owners start buying receivers. The civilian customers of Loran quickly outnumbered military ones. This led to expansion:
a. into a set of interlocking chains around CONUS intended to reach from beach to 100NM offshore … so incoming ship can make accurate landfall. A side effect was that 2/3 of land area of US is also covered.
b. FAA finally, after several wannabe air nav systems, agreed to support the coverage of the remaining US (two additional chains — pretty economical).
2. DoD uptake on Loran C, especially in the Navy, was spotty and irregular. As late as 1990, some classes of surface ships lacked Loran receivers (and integration into the nav problem). Some Loran equipped aircraft (e.g. P-3) had Loran receivers all right, but they were ancient trace-match manual ones. And, as noted above, parts of the Navy were trying to reinvent their own.
It appears, in retrospect, that the Navy tasking USCG for the boomer nav support was exactly the right thing to do. How would you institutionalize that kind of mechanism in any acquisition program?
Marv, Typically short tour lengths for PMs contributes to additional confusion, delay and cost. The acquisition process for high value systems is complex, burdened with an excess of checkers and gatekeepers. Dealing with the profusion of distractions these well meaning folks engender is best done by very experienced PMs. Short tour, ticket punch detailing for PMs exacerbates the situation.
Well stated Millard, another reason our great PMs made a difference… because they were in place for long periods of time… thanks
Unfortunately, the current situation is unlike anything anyone has confronted previously All of the DoD acquisition reforms and regulations may be well along the process of becoming completely irrelevant. When executed in an optimal manner, which is very difficult to do precisely because of cultural and regulatory problems, hence the need to pursue alt methods, AI systems have the capacity to exponentially improve the rate of R&D. It hasn’t yet fully cracked the wall of denial across incumbent R&D institutions. China is moving very rapidly while the U.S. debates another round of reforms, and funds the same old post war institutions and companies in much the same way. The institutions have never faced the combination of a technology and adversary like this before. I just don’t see much evidence that the incumbent DoD ecosystem is competitive — not in decision making, cost management or speed.
Another thought provoking article. As DOD seeks to embrace Agile Acquisition (AAM) and Technology Business Management (TBM) standards of practice already proven out in the market, we see other impediments undermining well meaning efforts like Kessel Run, EITaaS, DOD’s Cloud Strategy and other IT Modernization initiative, PMs will need to retool their existing SDLC and Sourcing processes that move away from a compliance and check box mentality, to one that is outcome and data driven; Business Value, Lifecycle Cost and Time to Market. Concepts of Decision Analytics and Value Chain Analysis should be applied to all IT modernization efforts, which will clarify stake holder inputs/outputs and expected time frames. IT-Acquisition Advisory Council (IT-AAC) embodies this work and provides agencies with the “penicillin for what ails Defense IT”. The patient will also need to move away from the butts in seats model of old guard contractors whom profit from implementations of these antiquated processes.
Thanks John, I couldn’t agree more that acquisition programs should be outcome driven, not compliance driven. Unfortunately the large bureaucratic processes leave little time for anything else but compliance time wasting processes.