Abstract:
From a military operations perspective, the primary goals of the Modular Open Systems Approach (MOSA) [Ref. 1] are Affordable Life Cycle Cost (ALC), Maintain Capability Overmatch (MCO), and Faster-to-Field (FtF). Achieving these goals is directly related to the ability of the architecture to enable reuse and interoperability. The importance of this association is amplified with the assumption of a Family of System (FoS) acquisition where mission-level capabilities are needed by multiple products. Accordingly, the value of MOSA to architecture development in a FoS acquisition is directly related to the ability of the resulting architecture to enable reuse and interoperability of capabilities across the products that comprise the FoS. Not only is the resulting architecture a key enabler, but the governance of that architecture is also critical to long-term success for a FoS with products that are not on the exact same development timeline. Model-Based Engineering (MBE) within a digital thread offers structures and processes that assist both architecture development and governance while integrating the best features of Model-Based Systems Engineering (MBSE) [Ref. 2], DevSecOps [Ref. 3, 4, 5, 6], open system standards, and Agile methodologies [Ref. 7]. However, the usage of such structures and processes needs to be done carefully to avoid situations in which the process itself can enable the intended or unintended consequence of sacrificing the benefits of MOSA while implementing the process. Without the right discipline, even developers with the best intentions for achieving MOSA benefits can end up building architectures, components, and build processes that inhibit the realization of MOSA benefits and complicate governance to an intractable level [Ref. 8]. Another concern is organizations that actually seek flaws in the development or governance processes and use those flaws to claim compliance with MOSA while at the same time embedding locks to guarantee future business contrary to MOSA goals. This paper will identify the discipline needed to achieve the benefits of MOSA in a FoS acquisition and recommendations for architecture development and life cycle governance. Building upon previous work [Ref. 9], this paper will present use cases and examples of pitfalls to avoid with suggestions on how to avoid them. The most significant recommendation provided in this paper is the development of a component-based architecture to be used as a basis to identify opportunities for reuse and interoperability across products and how to use it to establish a governance framework for an FoS-based acquisition. This paper describes the steps to derive a component-based architecture starting with a modeled reference architecture. Derivation follows the pattern endorsed by the US Government in their Comprehensive Architecture Strategy [Ref. 10] and includes process steps for iterative top-down decomposition and bottom-up re-composition. This paper presents study results on applying these principles against an example reference architecture. A digital map component is used as a representative example to highlight the pitfalls of reverse engineering a component architecture from an a priori system and a specific component implementation. Results will support the hypotheses that: (1) use of a reference architecture supports the creation of a baseline, point of departure, set of components aligned to an FoS product line, and (2) governance at the component level is best for a FoS acquisition.