From model-based requirements to an intelligent systems engineering advisor that identifies gaps in requirements: An application to space systems

Requirements have traditionally had a prominent role in the development of space systems. Engineers pursue completeness when establishing sets of requirements to avoid inadvertently developing or accepting systems that are present interoperability problems or a simply not fit for purpose. Yet, requirement gaps are common in the development of space systems. If lucky, those gaps are identified (and costly corrected) before the space system is deployed. Otherwise, those gaps reduce the success of the space mission as the system becomes operational. Unfortunately, the completeness of a set of requirements cannot be demonstrated; rather, completeness is a pursuit. Hence, the level of comprehensiveness of a set of requirements that can be attained is heavily driven by the engineer or team put to the task of defining the problem space. In this paper, we present a proof-of-concept of a virtual systems engineering (SE) advisor that supports the engineer in identifying gaps in the sets of requirements. The SE advisor evaluates requirements that are formulated as models and leverages a knowledge repository to reason about the meaning of the model-based requirements. If potential gaps in the set of requirements are identified, then the SE advisor presents them to the engineer, who decides how to address the gaps. For the proof-of-concept presented in this paper, requirements of a notional communications payload were modeled using the True Model-Based Requirements (TMBR) framework, which was implemented using an extension of the Systems Modeling Language (SysML). The SE advisor was embedded as a plug-in in a Model-Based Systems Engineering (MBSE) environment. The knowledge repository was limited to a set of hardcoded rules that captured knowledge associated to structural relationships of the model. Specifically, the rules capture fundamental meaning of model objects instead of syntactic ones. In this way, the SE advisor is purposed towards assessing requirements validation, not verifying model construction.