A Design for Testability (DFT) strategy for the development of highly complex safety-critical system using a Model-Based Systems Engineering (MBSE) approach

The concept of Design for Testability (DFT) involves enhancing a system's design with functionalities and features that facilitate testing during development and lifecycle stages, including system design, implementation, integration, verification, validation, production, and testing. When employing a Model-Based Systems Engineering (MBSE) approach for system design, it becomes essential to formulate an MBSE methodology to define how the system will address DFT requirements and strategies, seamlessly integrating them into the overall system design model. This work proposes an MBSE methodology that defines a DFT approach where system functions and physical features planned for testing, operations, maintenance and support and their relations are captured in a systems model using the systems modeling language (SysML). While testability in software has undergone extensive study, often driven by error occurrences (induced error), complex safety-critical systems primarily comprised of highly coupled hardware call for a different approach. The proposed DFT methodology includes failure mode and effect analysis (FMEA) artifacts that allow addressing the current constraints in the identification of multiple failures, and the roll-up of these failures to other elements at the same level or a higher level. The work also defines DFT metrics based on the current state of practice and a thorough literature review. These DFT metrics are defined as measures of effectiveness and performance in the system model and are demonstrated with an example for the development of a common hydromechanical system. A comprehensive framework, approach, and methodology that enables engineers to perform DFT while they are still early in the engineering design phase are proposed, considering the development of a system, its validation test system, and production test system.