Does SE Need to be Modernized, and What Might That Look Like?
Background
Modern systems engineering (SE) evolved in the 1940s through 1960s as an approach to manage both technical and programmatic risk in large complex systems. Systems engineering principles and methods were adopted by the DoD in the late 1960s/early 1970s as a way to manage technical and programmatic development and risk across the engineering and management components of large complex weapon systems. When the first iteration of DoD 5000.01 “The Defense Acquisition System” was published in 1971, it defined an SE-related set of guidance, including consideration for problem/operational needs, alternatives, test and evaluation, support and update as well as contracting, risk, source selection, and documentation. Systems engineering has been a foundation of DoD acquisition policy since it was formally defined as a system itself.
Principal Investigator: Mr. Tom McDermott, Stevens Institute of Technology
Why Systems Engineering Modernization?
The discipline of SE and its use in DoD acquisition has long been associated with realization of physical systems and related equipment. Today many defense capabilities are not only physical; they are also software intensive, highly connected, and have extensive automation and user configuration capabilities. Software engineering became a discipline in 1967, manufacturing automation (the third industrial revolution) began in the 1970s, and the World-Wide-Web was invented in 1989. The DoD’s Defense Modeling and Simulation Office was opened in the early 1990s and large-scale networked simulation of defense systems followed. All of these events have continued to evolve the discipline of systems engineering, not as a whole, but as a set of related subdisciplines (software systems engineering, information technology and enterprise architecture, distributed modeling & simulation, and automated manufacturing systems).
Following successful evolution of the Unified Modeling Language (UML) in the software discipline, the Systems Modeling Language (SysML) was published in 2007 and started the growth in Model-Based Systems Engineering (MBSE) as an improved approach to manage technical and programmatic risk. “Industry 4.0” originated in 2011 and introduced the concept of a “digital twin” as a non-physical product realization. The DoD’s Digital Engineering (DE) Strategy was published in 2018, ushering in the digital era of systems engineering. As the International Council on Systems Engineering (INCOSE) noted in their Vision 2035 document: “The future of Systems Engineering is Model Based, leveraging next generation modeling, simulation and visualization environments powered by the global digital transformation, to specify, analyze, design, and verify systems.”
Throughout all of this change, the “mainstay” of SE, and associated DoD acquisition guidance, continued to center on physical realization of large-scale monolithic weapon systems and other critical capabilities intended to persist for many years. The need for rigorous definition, analysis and testing of these critical systems will always exist, but the time has come to reintegrate the systems engineering subdisciplines into a common framework that responds to the digital age.
In 2021, the DoD published the latest 5000 series guidance, “The Adaptive Acquisition Framework” (AAF), which recognizes new development and acquisition pathways for software, IT and business systems, services, and a streamlined “middle tier” acquisition for more mature rapidly fielded systems. This followed a series of legislative directions to the DoD-tied four focus areas:
- Digital Engineering (DE) – An integrated digital approach that uses authoritative sources of system data and models as a continuum across disciplines to support lifecycle activities from concept through disposal. Digital Engineering will provide for the development, validation, use, curation, and maintenance of technically accurate digital systems, models of systems, subsystems, and their components, at the appropriate level of fidelity to ensure that test activities adequately simulate the environment in which a system will be deployed.
- Modular Open Systems Approach (MOSA) – An acquisition and design strategy consisting of a technical architecture that adopts open standards and supports a modular, loosely coupled and highly cohesive system structure. This modular open architecture includes publishing of key interfaces within the system and relevant design disclosure. MOSA introduces the “build for change, not to last” philosophy from software architecture across all aspects of DoD systems.
- Mission Engineering (ME) – The deliberate planning, analyzing, organizing, and integrating of current and emerging operational and system capabilities to achieve desired mission effects. Mission Engineering is intended to provide engineered mission-based outputs to the requirements process, guide prototypes, provide design options, and inform investment decisions.
- Agile Development – Approaches based on iterative development, frequent inspection and adaptation, and incremental deliveries, in which requirements and solutions evolve through collaboration in cross‐functional teams and through continuous stakeholder feedback. Agile approaches begin not with detailed requirements, but with a high-level capture of business and technical needs that provides enough information to define the software solution space, while also considering associated quality needs (such as security).
Program managers, contracting officers, and others now face a myriad of acquisition process changes centered on this need for more rapid deployment of capabilities, better weapon system portfolio management, and efficiencies created through digital transformation. This combination of factors brings increased complexity and risk to the system development lifecycle. There is a need for documentation of lessons learned, program best practices, and standard guidance for program SE that incorporates a holistic approach, and that approach can include many of the modernization activities mentioned. In addition, this guidance needs to apply to all acquisition pathways. Each pathway will require different approaches to program SE activities and associated guidance for those activities.
The SE Modernization project has three primary goals: 1) build an integrating framework that incorporates key activities across these domains and focus areas; 2) align and integrate these systems engineering practices to specific acquisition pathways; and 3) develop a set of artifacts and associated meta-data for a categorization and information framework that captures policy, guidance, and lessons learned into a body of knowledge.
The SERC team began addressing this task by conducting initial analyses of legislation and other policy documents related to the modernization focus areas. These analyses revealed two important findings: there is little integration across these focus areas in individual policies, and the intent of the policies as stated in the definitions above could be improved. For example, policies on agile do not generally reference DE – even though implementation of DE could be an enabler for agile.
As the team developed the integration framework, they came to realize first that existing SE lifecycle models like the “Vee” model and the DoD’s “Defense Acquisition Wall Chart” do not promote the future vision of data and models at the core of SE. Secondly, since future systems will be “built for change” using concepts of continuous iterative development, it was considered whether the somewhat linear models of existing SE lifecycle representations were still adequate guides. In response, the team shifted to developing a new conceptual view of the SE Modernization Lifecycle:
This view is complex but with study it becomes insightful in several ways. First, it illustrates systems engineering as a cyclic rather than linear approach. Though almost all literature attempting to standardize on a lifecycle model will say that activities are ongoing and should continue through the lifecycle, the circular illustration drives this point home differently. The view incorporates traditional acquisition milestones (triangles in Figure 2) but highlights them in the context of the multi-faceted work going on and where they fall within the broader context. Second, this view emphasizes the DE transformation using a layered model with data storage and transformation at the core, models as the data transformation layer, and systems engineering process areas as the outer layers. Perhaps most insightful are the colors of the outer ring as related to “Build/Measure/Learn,” which demonstrates the underlying goal in many of the new acquisition pathways for continuous iterative development of warfighter capabilities. For example, Prototyping, Agile, DevSecOps, and Lean are all continuous iterative approaches, though that language has not come through clearly in the existing policies. The view of a modern systems engineering approach overlaid with acquisition helps to bring this critical component forward.
The SE Modernization Lifecycle is still a work in progress, and the team will continue to iterate based on further research and engagement with the Defense acquisition and SE communities. The Principal Investigator for SE Modernization, Tom McDermott, SERC Deputy Director and CTO, will lead a strategy session for the International Council on Systems Engineering (INCOSE) in Summer 2022 to gain critical feedback and insights from the broader SE community. For additional detail, or for information on participating in the strategy session, contact Mr. McDermott.