BY: Ki Lee
Mar 11, 2022 | 9:00 am
The Department of Defense (DoD) last month released its Software Modernization Strategy, an important step to unifying existing technology and directing a more joint approach to systems of the future. It notes that our competitive posture is “reliant on strategic insight, proactive innovation, and effective technology integration enabled through software capabilities.” Further, it asks DoD entities that are driving software development to address management and governance of the 29 software factories spread across the services.
The promise of software factories is significant. The construct has quickly demonstrated agile development of critical capabilities, while also delivering the speed and flexibility that DoD requires. This has been achieved with commitment to automation, modular and open architecture, and continuous authority to operate (cATO). Given how much progress has been achieved, software factories will undoubtedly play an important role in modernizing DoD technology, through a common framework which will drive efficiencies and provide oversight, best practices, and baseline software for reuse.
The industry can offer lessons learned and best practices from supporting software factories over the past five years, and the Software Modernization Strategy makes several mentions of the role of industry within the acquisition process. Booz Allen has worked in partnership with the DoD to rapidly develop, integrate and field cutting edge mission capabilities. We know what works, and what it will take to mature the software factory ecosystem to bring more efficient, innovative software to the DoD.
The software factory ecosystem cannot be a one-size-fits-all approach. Given different mission requirements (i.e., from IT to OT), acquisition strategy and program maturity (i.e., inception to sustainment), bringing all of these software factories into a single factory or even entity will likely be both inefficient and ineffective.
DoD leaders would benefit from establishing an ecosystem that streamlines across today’s software factories and those that will exist in the future. We would offer that the DoD Chief Information Office should consider the following as a construct aligning software factories within a framework for governance:
Mission Purpose. Each software factory was purposefully designed to cover a broad spectrum of mission requirements. Some entities support complex combat requirements and require deep hardware integration for use on the battlefield driving additional layers of policy, cyber, security and technical specifications. Others focus on IT system development that requires less hardware integration, but instead requires standards for open architecture and system interoperability. Aligning against a core set of mission categories will help streamline baseline software and regulations.
Developmental Maturity of System. DoD must also consider the maturity of the software that’s being developed in applying governance to the software factories. For instance, a software factory for a system as mature as the F-35 program has very different needs (e.g., cyber regulations) than an entity that is incubating new software for prototyping and testing capabilities with more pipeline requirements (e.g., Navy’s Rapid Autonomy Integration Lab (RAIL) software factory).
Operations Model. Finally, the framework should address the variety of different models of ownership and accountability. Depending on how the DoD and industry partner together, the risks and benefits can vary significantly. In our experience, there are three broad models for operations that work effectively for managing cost, risk, and innovation: Government-Owned, Government-Operated (GoGo); Government-Owned, Contractor-Operated (GoCo); and Contractor-Owned, Contractor-Operated (CoCo). We have supported all of these models across the federal government and have lessons learned from each, including retaining government data rights, ensuring accountability and driving a culture of collaboration.
As the DoD works on the implementation plan for the Software Modernization Strategy and creates a framework for how to manage and govern its software factory ecosystem, aligning and categorizing its existing resources will be critical for delivering agile, mission critical software to current and future weapons systems.
Further, it is incumbent on the industry to continue to streamline and drive efficiencies in its software development. Across the Defense Industrial Base, we must commit to building open, reusable baseline software that can be extended and augmented across multiple use cases. Whether we are developing software on corporate research and development budgets or on government use cases, creating software that can be refined to meet mission needs is absolutely critical. From our own experience, we are committed to taking our investments and integrating them to build for custom mission solutions. Our AI/ ML software factories and cyber software factories are built consistently from a horizontal perspective so that they can be leveraged and reapplied for more vertical mission sets.
Through both a more consistent government framework for software factories built on real use cases and industry commitment to re-baselining, executing this new strategy will be critical for continuing to modernize technology to support today and tomorrow’s warfighter.