作者:
Brown, GregJain, AnandNI
Offering Management & Business Dev Austin TX 78759 USA NI
Res & Dev Austin TX USA
model-based test engineering (MBTE) is an approach to how test can deliver new, unique, and increased value to organizations and programs in the Digital engineering landscape. Using standards such as SysML[1] and ATML...
详细信息
ISBN:
(纸本)9798350300284
model-based test engineering (MBTE) is an approach to how test can deliver new, unique, and increased value to organizations and programs in the Digital engineering landscape. Using standards such as SysML[1] and ATML[2], a common language (style guide) and set of workflows are being defined that bridge the worlds of model-based System engineering, Design engineering, and test. There are many challenges that exist today that increase the costs and schedules of programs. These include requirements sufficiency, tracking requirements across the program life cycle, knowledge capture, lack of "authoritative sources of truth (ASOT)," TPS development and maintenance, and ATE maintenance and obsolescence. This paper provides an overview of MBTE, what needs to be described in the SysML domain and how that enables high value such as keeping the SysML models as the ASOT for test and automating the downstream ATE workflows such as TPS generation. MBTE enables Systems engineering, Design engineering, and test to have a common language to achieve greater understanding and agreement as to what needs to be tested and improve decision making. With so much investment being put into models, extracting the most value out of those models is imperative. MBTE is an approach that is being built to extract that value across the program life cycle.
model-based Systems engineering (MBSE) and Digital engineering (DE) are poised to significantly transform the automated test sector. However, the broad integration of these principles is hindered by issues related to ...
详细信息
ISBN:
(纸本)9798350349436;9798350349443
model-based Systems engineering (MBSE) and Digital engineering (DE) are poised to significantly transform the automated test sector. However, the broad integration of these principles is hindered by issues related to the completeness and uniformity of existing technical data. Organizations responsible for automated test systems are often dependent of a diverse array of test requirement documents, ranging from military to contractor standards, with some legacy documents not adhering to any recognized standard or being entirely unavailable. There is a pressing need for mechanisms that can facilitate the transition from these varied legacy formats to contemporary digital engineering tools. This paper proposes a comprehensive strategy and a tooling framework designed to convert a wide range of legacy documents into a unified XML-basedmodel format. It outlines various use cases and tooling frameworks, along with their foundational methodologies. Additionally, the paper examines the effects on the standard timelines for test Program Set Development and proposes a structure for future innovations that leverage machine learning and artificial intelligence to address some of the sector's most challenging issues. Leveraging artificial intelligence is a key enabler for many of the "legacy document to XML" translation tools presented. Modern Large Language models (LLM) can recognize patterns in test requirements and discern meaning from the placement of data within a table or flow chart. This is powerful because it can drastically improve the quality and speed at which legacy documents are converted from PDF format to the model-based eXtensible Markup Language (XML) formats like IEEE ATML 1671. When test requirements are clearly defined upfront and not hidden from the engineering organization in legacy documents engineering managers, customers and developers can all better support the end users with accurate and faster test Program Set development. With models in hand test ass
暂无评论