Shifting Left: A Test-Integrated System Modeling Approach to Reduce System Development Cycle Time

Awele Anyanhun (Georgia Tech Research Institute)

Keywords
Model-Based Systems Engineering;Model-Based Test and Evaluation;Shift Left;MB-Testing;System Development Cycle Time
Abstract

Problem Discovery during operational test and evaluation (T&E) has a significant impact on the acquisition cycle time of systems. A loss in competitiveness in the marketplace due to delays in fielding systems or products can prove detrimental to the survivability of organizations, and is a critical risk that must be actively monitored and mitigated. To achieve a reduction in the acquisition cycle time, i.e., a Shift Left, for a given system, traditional methods of testing need to be modified to accommodate more systemic and disciplined approaches that consider T&E more as a continuum, beginning at the conceptual phase of system design and development. Consequently, a systems engineering approach that introduces test capability modeling as an integral part of the model-based systems modeling methodology would result in a fully test-integrated MBSE approach to descriptive system modeling that emphasizes alongside the system-of-interest’s (SOI) architecture, a test capability architecture, i.e., the necessary test resource artifacts required for verification and validation.

This proposed approach differs from most Model-based T&E approaches reviewed in literature due to the specification of testing capabilities as an inherent part of a model-based test context and configuration. Most traditional model-based approaches usually capture a tester model element within the test context as the model element initiating a given test case. However, our approach specifies within a test capability model, the test facility and test resources required to perform testing of the system and/or specific capabilities as part of the test context. This approach accomplishes several T&E goals: firstly, it alerts the system architects and decision makers to the suitability of a specific test facility’s ability to test to the system requirements and/or operational capabilities due to the availability of either adequate or inadequate test resources.

Secondly, the approach utilizes a model-based test environment which could significantly reduce the overall T&E costs for the actual system by enabling the exploration of multiple test case specifications using testing techniques such as simulation via model execution, and analysis in order to identify the right tests and methods that could be deployed in testing the actual system. Performing verification and validation of a system using system models is a cost-effective way to show theoretical compliance of the system architecture prior development. Accordingly, in this work we demonstrate how a model-based test integrated configuration can improve a system’s acquisition cycle time with a specific use case. A notional Missile System constitutes the SOI/SUT for our exemplar and is modeled using SysML. The notional test facility, i.e., the test capability model is defined as a model-based representation of required test resources ― including representations of the system’s operational environment ― required to enable testing of a given set of system capabilities.