Article targeted at: systems engineers, quality managers, architects, quality assurance teams.
Executive Summary
For years, Enterprise Architect has stood out as the go-to modeling tool for automotive and other (medical, mission-critical, business etc.) ventures. Now, with the introduction of the IncQuery Validator for Enterprise Architect, the duo aims to revolutionize the modelling standards. This sophisticated addition extends beyond mere model validation, offering a comprehensive approach towards full-fledged quality governance. With its blend of innate and customizable rules, IncQuery Validator ensures stringent adherence to both ISO26262, Automotive Spice and other modelling guidelines. Quality assurance engineers will appreciate its ability to not only provide a high-level Quality Score but also deliver meticulous breakdowns of quality-related findings. This in-depth insight allows them to precisely evaluate metrics and confirm adherence to both standard as well as custom rules, ensuring each project is on the right track. Furthermore, the tool equips other stakeholders with invaluable evolution insights, aiding in strategic decision-making. The IncQuery Validator for Enterprise Architect is set to become an indispensable asset in model-based systems engineering and quality assurance.
Introduction
IncQuery provides a new platform for digital engineering, that consists of three key elements:
- Handover Automation to boost productivity and eliminate document-centric workflows;
- Automated Quality Gates to ensure engineering work free of design errors;
- Digital Thread Analytics to break the vendor and data lock-in.
As a result, our customers may average double engineering productivity and reduce costs by 30%. In this article we would like to focus on Automated Quality Gates and Quality Governance.
Key pain points for introducing quality governance
Along the steps of the safety- and mission-critical engineering V model, processes tend to create disconnected data silos, miscommunication, and duplicate work. A typical silo can be an MBSE tool, MS Excel, or detailed design tools (domain specific, such as mechanical, electrical etc.), all difficult to integrate with the rest. Loads of document-centric work would happen. Interfaces between domains are often documents, such as PDF, Word, Excel, etc. What is in the transfer documentation is snapshot, may be disconnected from the original source of truth (silo). If you want to maintain consistency, a lot of manual work needs to be done. Data re-entry and copy-pasting happens, with loads of time and money wasted. As a result, we lose the guarantee of consistency.
That problem brings us why we are here today. How to guarantee consistency through Model Validation and on a bigger scale, Model Governance.
Enterprise Architect has some basic quality-related built-in features: performance rules, element composition rules and property validity rules. These features were designed to ensure authoring success and may free users up from fundamental errors (errors for instance: creating a diagram with an element that should not be there, same ID for two elements), but running validation for large projects is often slow - , which results in a lot of people just skipping running validation altogether. The organization comes across quality gaps where projects are plagued by numerous issues like incorrect profile applications, missing attribute values, non-standard-compliant models, design errors, or anti-patterns. As they accumulate over time, they become a major productivity hindrance and source of missed deadlines and cost overruns.
Model Validation and Model Governance in Enterprise Architect
We at IncQuery focus on guaranteeing consistency, by extending the practice of Model Validation and Model Governance.
- Model Validation: quality criteria that define well-formedness conditions to ensure authoring success. Validation focuses on a single SysML project. Goal is to help systems engineers to assess their key quality related metrics of their, independently on what environment they work. (Enterprise Architect, Catia, IBM, Genesys)
- Model Governance: quality criteria that define the preconditions for data exchange. Enforce policies (style guidelines) across several stakeholders (organizations). Governance focuses on multiple projects, even repositories. Goal is to help downstream stakeholders, who consume the modelling outputs, quality insurance experts, SW architects, etc. Model Governance is key to achieve consistency across several disciplines. If one makes/maintains electrical design based on system model architecture, it is important that all artifacts, document that she bases her work on is compliant with her quality requirements.
Features of IncQuery Validator
- Can run as an add on or standalone application or as part of a DevOps pipeline (e.g., Jenkins).
- Comes with built-in rule libraries covering basic check for UML and SysML standards conformance, profile stereotype application correctness, and basic sanity checks for diagrams.
- Can perform in-depth design error and anti-pattern analysis based on well know libraries such as the SAIC Digital Engineering Validation ruleset, or SysMOD
- Provides a convenient extension framework to define custom validation rules for models.
- Supports centrally shared & version managed projects by integrating with files-based VCs such as Git/SVN.
- Provides a quality score, high-level KPI for overall quality status.
- Creates rich HTML/PDF or MS Excel csv reports so you can share and review analysis reports using popular collaboration platforms like Jupyter and Confluence. (see below figures)
The concept of Digital Engineering governance in terms of DevOps: in Software Engineering, DevOps pipelines provide a way of automating different operations: building, compiling, testing, analyzing - using an automation platform. The same principle can be applied to MBSE, to bring collaborative systems engineering and in general, modeling to the next level.
Customizing validation rules
IncQuery Validator rulesets are written in the Viatra Query Language (VQL). There are three options to establish a ruleset, based on the need:
- simpler rules: systems engineers can create after a shorter VQL training from IncQuery,
- comprehensive rules: IncQuery offers a one-week training on VQL,
- complex rules: IncQuery may decide if option 2 is satisfactory or initiating a consulting project would be necessary.
Conclusion
The IncQuery Validator is an automated quality gate which provides quality reports that can be based on MBSE standards and custom defined modelling rules. The product provides Systems Engineers and Quality Managers with capabilities to assess quality metrics, and downstream stakeholders to assess quality of inbound models.
- Validation à Governance
- Quality criteria for data exchange.
- Automated Quality Gates
- Enforce guidelines, styles and quality metrics across stakeholders and organizations.
- IncQuery Validator
- Tool-independent rules
- Automated execution via DevOps pipelines
- Integrated with reporting and dashboards via rich HTML and Excel/CSV
- Quality Score as a high-level quality KPI
Useful links
On IncQuery Validator for Enterprise Architect (product page): https://incquery.io/validator
On the Enterprise Architect-based full ecosystem: https://www.youtube.com/watch?v=CuUvoBPNY3Y
On Viatra Query Language (product page): https://eclipse.dev/viatra/documentation/query-language.html
IncQuery Group: www.incquery.io
About IncQuery Group
Vienna, Austria based IncQuery Group is a technology company specializing in model-based software engineering and artificial intelligence solutions. Their expertise lies in developing innovative tools and techniques to help businesses improve their software development processes and create reliable, high-quality software development toolchains. With a focus on model-based systems engineering, they empower organizations to enhance productivity and reduce development costs. IncQuery Labs is dedicated to advancing the field of software engineering through innovative research and practical, industry-focused applications.
The format of each course will be varied, depending on the subject, audience and location; sessions being held in July feature a mix of both online and face-to-face seminars and courses.
