This is Part 2 of a blog series covering a whitepaper titled,The Comprehensive Guide to Successfully Adopting Model-Based Systems Engineering MBSE. Visit these links for the rest of this series: Part I, Part III, and Part IV.
What does it mean to practice SE from a data-centric perspective?
Successfully adopting MBSE and moving toward a data-centric practice of SE is much more than just acquiring and using a specific tool, set of tools, or focusing on the use of a specific type of model. As stated previously, MBSE is not just about the development of SysML or other language-based models nor just practicing model-based design. MBSE is itself made up of puzzle pieces, all of which contribute to the successful adoption of MBSE. To be successful, the following ten areas of capability associated with data-centricity must be addressed.
01: Holistic Product Development
A key tenet of data-centricity is taking a holistic view of product development and managing data and information within an integrated/federated environment. The focus is on multidiscipline, collaborative, project teams (e.g., integrated product teams). Many organizations still operate in organizational silos, with team members’ loyalty toward their specific silo rather than to the project team. When issues occur, the tendency is to blame those in other silos. Each silo often has its own processes, specific toolsets, data, and artifacts. Often the data and information are generated independently from those in other silos and are not in a form to enable sharing. This can result in inconsistencies, correctness, completeness, and currency issues of the data maintained in these artifacts. When moving toward data-centricity, organizations must have a holistic view of product development, minimizing the silos, encouraging collaboration, and improving communications not only between team members but between different tools used to generate and maintain data and information. Rather than treating Systems Engineering separate from Project Management (PM), projects must integrate both sets of functions such that there is a single project team that does both functions.
02: Manage Product Development Across the Lifecycle
Rather than having tools that are specific to a given organizational silo, a key characteristic of data centricity it that related data and information that represents lifecycle activities and associated artifacts can be linked resulting in “digital threads” that link related information together across the product lifecycle. This linkage enables project team members to work collaboratively and establish traceability between needs, design input requirements, system analysis artifacts, diagrams, models, architecture, design, system verification artifacts, and system validation artifacts. Traceability aids in change impact assessment across the product lifecycle helping ensure completeness, correctness, consistency, and currency of the data and information and resulting artifacts.
03: Enterprise Level Data and Governance Policy, Processes, & Procedures
Because of the dependence of not just the project teams, but the overall organization on electronic forms of data and information and increasing threats associated with the security of this data and information; enterprise-level policies, processes, and procedures concerning data governance and information management must be defined, implemented, and enforced.
04: Project Level Data and Information Management
Within the context of the enterprise-level data governance and information policies, each project must include their specific implementation of these policies within their Project Management Plan (PMP) and Systems Engineering Management Plan (SEMP). Because of the importance of managing the project’s data and information, the project is encouraged to develop and enforce a project-level Information Management Plan (IMP). Other supporting plans (e.g., requirements management plan) need to comply with the data management policies within the higher-level plans for both the project and enterprise.
RELATED POST: The Real Intent of Model-Based Systems Engineering
05: Master Ontology
Terminology and language are key to successful communications not only between team members but between tools. For a given enterprise, an enterprise-level ontology (data dictionary and glossary) must be developed to clearly define specific terminology and relationships of various terms used within the organization and the project. This is critical when there are product lines, multiple project teams, and the need to share data and information between current projects as well as reuse data and information for future projects. Within the enterprise-level ontology, individual project teams can define their project-specific ontology consistent with the enterprise-level ontology.
06: Master Schema
Here the word “schema” is used to describe how the data and information are organized and managed within individual tools and associated databases. It includes the naming of individual data and information items, defining relationships between data items, and the import and export of data and information. From both an enterprise and project perspective it is important to define a master schema that the SE and PM tools within their toolset are compliant in order to enable data integration, shareability, and reuse.
07: Use of Attributes and Associated Measures
Data centricity enables the project to define and use attributes that can be used to manage project activities across all system life cycle stages. For needs and requirements, attributes can include rationale, priority, criticality, source, owner, traceability, risk, maturity of needs definition, needs and requirements definition status, design implementation, system verification, and system validation. Attributes can be defined to aid in reusability and product line management. Attributes can also be associated with key measures defined by stakeholders within their goals and objectives. These measures include key performance indicators (KPI), measures of suitability (MOS), measures of effectiveness (MOE), measures of performance (MOP), key performance measures KPP), technical performance measures (TPM), and leading indicators (LI). Data representing these measures and attributes can be used within the SE and PM tools to generate reports, dashboards, etc. which are used to better manage the project and system engineering processes providing managers near real-time status information and enabling them to identify and correct possible issues before they become problems.
08: Configuration Management
Adopting data-centricity, the project’s artifacts and underlying data and information are developed, analyzed, and managed holistically within the data and information model. Because the data and information are managed within the project’s data and information model, this model represents a single source of truth (SSoT) for the project. Rather than configuration control of each individual artifact represented by the data and information in the model, the project team can configuration control the model which represents the baseline state of the artifacts represented by the data and information in the model at any given time. “Visualizations” of the data and information in the form of the various artifacts represent the baseline version of that artifact. Even when these visualizations are extracted as reports, the SSoT is still the data and information model from which they were generated.
Note: for many organizations, this is often their biggest challenge in that it requires the organization to redefine its concept of configuration management. However, as stated previously, configuration management of individual artifacts requires significant overhead in both cost and time to individually configuration manage individual documents as compared to managing the data and information model that is representative of moving towards a data-centric practice of SE and PM.
RELATED POST: Webinar: Eliminating Barriers to MBSE Adoption with Jama Software
09: Systems Engineering (SE) Tool Set
Data centricity requires projects to move beyond the use of common office applications: word processing, spreadsheets, presentations, basic drawing and diagraming tools, and requirement only management tools to define, analyze, record, and manage needs and requirements and other SE artifacts. Rather, projects must transform their SE process such that SE artifacts are developed using SE tools that are fully compliant with interoperability and data sharing standards, are consistent with the enterprise and project ontology, stores the data and information consistent with the project’s master schema, and allows linking of data and information across lifecycle activities and resulting artifacts. This data and information must be managed in a form that is shareable between the SE tools within the project’s toolset as well as shareable with the project’s PM tools. When selecting specific SE tools to be included in the project’s toolset, it is important that the project determine the types of information and methods of analysis that are needed based on their specific product line, culture, and workforce.
10: Project Management (PM) Tool Set
Data centricity also requires projects to move beyond the use of common office applications for project management e.g., budgeting, scheduling, cost management, risk management). Rather, projects must transform their PM process such that most of the PM artifacts are being developed using PM tools that are fully compliant with interoperability and data sharing standards, are consistent with the enterprise and project ontology, stores the data and information consistent with the project’s master schema, and allows linking of data and information across lifecycle activities and resulting artifacts. This data and information must be managed in a form that is shareable between the PM tools within the project’s toolset as well as shareable with the project’s SE tools. For example, Work Breakdown Structures (WBS) can be linked to Product Breakdown Structures (PBS) and physical architectures to enable management of budgets, schedules, resources, and risks associated with each system and system element within the product physical architecture.
Stay tuned for Parts III through IV for more tips on successful MBSE adoption.
To download the entire paper, visit: Whitepaper: The Comprehensive Guide to Successfully Adopting Model-Based Systems Engineering MBSE