From the halls of the Pentagon to the individual fighting position at the tactical edge, there is not a sailor or Marine who has not lamented the challenges in developing, procuring, and fielding new capabilities in a timely and cost-effective manner. The acceleration of technology advancement is outpacing the ability to get new capabilities into the hands of warfighters before they are rendered obsolete or offset by an adversary.1
In the debate surrounding how to accelerate the Joint Capabilities Integration and Development System (JCIDS), the argument often devolves into a desire to change the law, modify the JCIDS process, or ask for greater authorities to overcome or bypass the perceived “frozen middle” of bureaucracy.2 However, making such changes would require the equivalent of another Goldwater-Nichols Act and would only address the symptoms, not the disease. The real problem is the lack of clarity in the roles and responsibilities across the entire acquisitions enterprise.
It is time to consider a new assembly line approach, in which there are clearly defined roles and responsibilities across the JCIDS. This assembly line starts with science and technology (S&T), proceeds to research and development (R&D), accelerates through test and evaluation (T&E), and then scales up through procurement and sustainment.
The Assembly Line Approach
In 1913, Henry Ford installed the first moving assembly line for the mass production of automobiles. His innovation reduced the time it took to build a vehicle from more than 12 hours to under 2.3 The Sea Services can apply a similar innovation to their overall capability development process.
Today, there are four different “assembly lines,” all of which overlap. These assembly lines move at different speeds, have unclear transition points, and lack an entity with the specific authority and responsibility (call it an assembly line manager) to manage the continuum of effort. In short, no one owns the proverbial “valley of death” (see Figure 1) and there is no oversight of the entire process to ensure duplication is minimized, responsibilities are clarified, and priorities are set.4
The S&T assembly line works from inception of an idea to a physical capability and typically ends with a technology readiness level (TRL) of around 5 or 6 (see Figure 2), but it falls short of a clearly defined transition process from this point.5
The innovation community line finds higher TRL 6 or 7 capabilities within the commercial and defense industry. It then attempts to inject these capabilities “mid-stream” into the planning, programming, budgeting, and execution (PPBE) process—by which resources are allocated among the military departments, defense agencies, and other components—usually with limited effect and no clear oversight or prioritization of effort.6 This assembly line is often disrupted by a lack of funding to transition into the five-year defense budget planning cycle—or worse, when there is no established program of record or requirement for a capability to fall into for procurement.
The requirements generation line builds capabilities-based assessments and provides requirements to the acquisition community, almost to the exclusion of S&T or the innovation communities. These requirements can be detailed through a series of capabilities design documents. Requirements are often so specific that they define a singular system or solution or require extensive development—frustrating industry because they are handed a specific request rather than asked to solve a problem.7 In addition, requirements are often built without insight into the S&T or innovation assembly lines and are only started once the requirements team is made aware.
The acquisition systems commands, program executive offices, and program management activities line covers almost the entire process, from “pre-milestone A” through procurement and sustainment, usually with significant overlap with the other assembly lines.8 It is not that they ignore the other assembly lines. Rather, it is that there is no codified process or clarity in terms of roles and responsibilities that directs the acquisition community to focus on transitioning capabilities from other assembly lines and then scaling them through procurement and sustainment.
In fact, under the current process, the requirements assembly line hands the acquisition community a requirement, instead of that requirement going to S&T or R&D for consideration. Then the acquisition community has statutory responsibility, called milestone decision authority (MDA), to conduct the analysis of alternatives, down-select a vendor, and then develop, test, evaluate, procure, and sustain that new capability.9 This often results in duplicative analyses of alternatives, industry studies, development of capabilities, and even complete redesigns from the ground up in the differing assembly lines.
An additional complication is that each program office has limited manpower. Therefore, if one office is responsible for developing a new capability from cradle to grave, while simultaneously being responsible for procurement and sustainment of existing capabilities, the team must bifurcate their focus and time. Every time that program office is given another new development program, they split their time even further. This problem exists mainly as a result of where Congress allocates money (into program elements) and who has overall responsibility. If an entity is given money, Congress expects that it will meet benchmarks in obligating and executing those funds.
These conflicting efforts add costs and increase schedule, all to reach a certain level of performance based on often overly specific and restrictive requirements. Add to this the differing risk calculi between S&T development, fleet operational urgency, and what is necessary to procure and sustain a program of record legally, and there is a collision of competing efforts.
This might seem like an insurmountable problem, but if the Sea Services can clarify these roles and responsibilities and consolidate funding within specific offices of primary responsibility, then the process could become a cohesive assembly line process with seamless transitions.
Requirements and Analysis
Currently, requirements are handed to the acquisition community, which then holds the milestone decision authority. Instead, requirements should be given first to S&T. This is where the JCIDS process should start. The well-funded S&T community should conduct the analysis of alternatives and industry studies and should canvas the S&T community to see if a solution already exists or is in development and, if so, at what TRL it resides. This analysis also should include close allies and partners to ensure there is no duplication. For the acquisition community to accept this analysis, sufficient rigor is necessary to meet its legal and statutory requirements for competition.
If the capability has a low TRL, it should remain with S&T for further development, leveraging the deep pockets of this community. The S&T community should be capable of making the material development decisions (MDD) up to the Milestone A decision, where the hand-off occurs for further development (see Figure 3).10
If the technology has a higher TRL, or when it is ready to move along the assembly line, it can encounter the “valley of death.” S&T efforts fail to cross this chasm for one simple reason: No one has been put in charge of the valley of death, nor has the RDT&E resourcing been consolidated here to emphasize transition from S&T, into RDT&E, and then over to acquisitions for procurement and sustainment.
This is the single most important change that needs to occur: The Sea Services need dedicated entities that consolidate these responsibilities, with the express purpose of transitioning capabilities across the valley of death, informing the requirements assembly line, forecasting when capabilities will be ready for transition, and capturing the innovation assembly line body of work. These should provide guidance, prioritization, and oversight to prevent duplication of effort. These offices should be backed up by the Under Secretary of Defense for Research and Engineering oversight and should be tightly aligned with both the combatant commands’ requirements and the Joint Requirements Oversight Council. They should focus on producing prototypes in the fleet, which in turn allows for developing tactics, techniques, and procedures, training plans, and new manpower models. These offices should conduct the technology maturation and risk reduction up to Milestone Decision B.11
At this point in the assembly line, it is important to talk about requirements generation. Requirements should enter the process at the beginning, with S&T. However, as a technology matures, the capability design document must be produced, and the requirement must be codified more clearly. If a top-level requirement was the genesis, it needs to be refined to more granularity. That said, it is imperative to avoid the trap of waiting for requirements to be refined to a 90 percent solution and for material solutions to meet 90 percent or greater of the stated requirement.
The rate of technological advancement calls for developing requirement “buckets” within which the services can iterate, instead of specific programs of record tied to single platforms or capabilities. Building programs of record and their associated requirements in a “system of systems” approach allows for greater flexibility as technology evolves. In addition, initial procurement and sustainment can be accomplished faster with the 70 percent solution, which is typically cheaper and gets some level of capability into the fleet’s hands sooner. This reduces pressure to meet operational risk requirements, and more time can be spent developing the full objective capability. The Marine Corps has demonstrated this approach with the rapid procurement and fielding of the MQ-9A Reaper, meeting its initial operational requirements inside 3 years instead of the 15 years it typically took to field an aviation program in the past.12 Now, inside the program of record, iterations will occur to grow the family of systems with other capabilities that address the full range of requirements.13
If the services could buy capabilities in “tranches” every couple of years in a continuous development cycle, instead of “buying in” to a 20- or 30-year program of record, the impact could be far reaching. It would encourage the defense industry to solve problems, knowing they would have a chance to recompete if they did not win a contract on the first attempt. It keeps the existing industry partner honest, because if they underbid on the contract to win the award, they could be replaced in the next few years. It also allows for the adoption of evolving technology, so the services do not get “stuck” with an obsolete capability when the threat gets a vote.
In this model, the requirement for a capability must be revisited every year to determine if modifications to objective requirements are necessary based on the ever-changing threat. It also drives faster replacement of technology. If a technology matures “midcycle” within the higher TRL innovation assembly line, it can be adopted more quickly, with an existing requirement bucket already there to catch it, with a simple requirements clarification letter.
The Final Step
The final step in this assembly line should be the acquisition community. If S&T were empowered to conduct the initial analysis of alternatives and development of capabilities, and if this were then handed off to a dedicated R&D/T&E entity, backed by a requirement that was prepared in parallel, the acquisition community would receive a body of work of sufficient quality to support the Milestone C decision.14
The most important point in this assembly line would be having the appropriate program office ready to catch the capability in stride. Leaders in the acquisition community would allocate programs transitioning out of R&D/T&E to the appropriate program office, so no one entity is overwhelmed with new starts while trying to sustain existing programs of record. Forecasting the sundown of existing programs would make room for replacements. This would be made easier if programs of record were “buckets,” with the ability to keep up with technological change.
Another opportunity in this process would be to establish “technology shelves,” where newly developed material solutions could be placed for the fleet to purchase with their own funding. This would go a long way to balancing the operational risk–acquisition risk equation and achieving the “speed to the fleet” the fleet desires, while alleviating pressure on the acquisition community to accomplish all the contracting work. A great example of this is small unmanned aerial systems, where the technology is changing so rapidly that it makes more sense to validate these capabilities and then put them on a tech shelf for the fleet to purchase as disposable end items.
For years, the Sea Services have been attempting to make do with the existing acquisitions methodology, and it is clear to all that they are falling behind. It is time to rethink the process.
1. L. Darina, “How Fast Is Technology Growing Statistics [Updated 2022],” Leftronic, 26 April 2022.
2. John Shanahan and Laura Junor, “We Need a Goldwater-Nichols Act for Emerging Technology,” Defense One, 16 December 2020; and “Project Management: The Incubator for Acquisition Reform,” Federal Times, 11 March 2016.
3. “Ford’s Assembly Line Starts Rolling,” History, updated 30 November 2021.
4. Peter Modigliani, “The Program Side of the Valley of Death,” Mitre AiDA, 17 March 2022.
6. Daisy Thornton, “DoD Confronting ‘Valley of Death,’ Other Innovation Bottlenecks,” Federal News Network, 14 June 2022.
7. C. Todd Lopez, “Over-Prescriptive Requirements Hinder Defense Innovation,” DoD News, 17 June 2020.
8. Title 10—Armed Forces, Subtitle A—General Military Law, Part IV-Service, Supply, and Property, 10 USC 2366a: Renumbered §4251.
9. “Acquisition Process: Milestone Decision Authority (MDA),” AcqNotes, 21 March 2021.
10. “Acquisition Process: Milestone Decision Authority (MDA),” AcqNotes.
11. “Acquisition Process: Milestone Decision Authority (MDA),” AcqNotes.
12. Scott Anthony Cuomo, “From No, to Yes, Maybe, and NIMBY: Explaining Variation in Remotely-Piloted Aircraft Adoption Between the U.S. Air Force and Marine Corps Since 1993,” thesis, Georgetown University, 2020.
13. Megan Eckstein, “Marines Ditch MUX Ship-Based Drone to Pursue Large Land-Based UAX, Smaller Shipboard Vehicle,” USNI News, 10 March 2020.
14. “Acquisition Process: Milestone Decision Authority (MDA),” AcqNotes.