Model-Based Systems Engineering (MBSE) is a powerful approach to managing complex system development, and STRATA MBSE methodology provides a structured framework for ensuring consistency, completeness, and correctness. But how does STRATA work in practice? In this blog, we’ll explore how to implement STRATA effectively, focusing on layered decomposition, cross-pillar alignment, and iterative refinement.
STRATA is all about decomposition and relation management. Many methodologies do not stress the decomposition of complex systems enough., Or, if they do, they don’t effectively manage the decomposition holistically. That’s truly the beauty of the STRATA methodology – it provides a sound framework for decomposition, simplifying the complexity into every finer pieces that can be understood on their own, all while maintaining all the relationships to the other elements across the requirements, behavioral, structural and verification domains. In short: a simplified, holistic, and effective capture of system knowledge.
Step 1: Establishing the System Model
Before diving into STRATA, it’s essential to define the system scope and establish an initial structured model. This involves:
- Identifying Stakeholder Needs – Capturing the high-level requirements and constraints from your customer. If possible, start with a capture and analysis of the original stakeholder’s need statements – the requirements. Look for issues and concerns with them.Have difficult conversations with your customer as soon as possible to clarify the requirements and ensure their feasibility.
- Defining System Boundaries – Establishing what is inside and outside the system. This becomes a basis for architecture and interfaces. Understanding where your system begins and ends is vital to forming your physical architecture and managing the interfaces. In today’s complex system development projects interface management is key to performing, as we all know systems tend to break down at the interfaces. The more we can do upfront to understand, clarify and properly communicate the architecture at every level and the interfaces, the more robust our systems become.
- Selecting the Right Modeling Language – Using CSDL and/or SysML to structure the model. Enforce the language across the program, ensure everyone is using the same terms and relationships to convey meaning that’s easily understood by all the engineering disciplines.
- Choosing the Right Tool – Implementing GENESYS or another full-featured modeling tool that recognizes system decomposition to manage relationships and dependencies. A good tool is essential to implement the methodology effectively.
STRATA allows engineers to start with what they know, regardless of where it fits in the system hierarchy. This flexibility ensures that early modeling efforts remain valuable as the system evolves.
Start the process with modeling the information you have. For example, if you have your basic systems logical hierarchy already defined down to a subsystem level, build the System Layer (Layer 1) structural architecture, and build your Subsystem Layer (Layer 2) structural architecture, then interconnect layer 1 and layer 2 with the proper hierarchical relations for both your components and linkages. This can serve as a starting point and is quite often the most common starting point.
At this point, follow the process in each layer, and derive the functional architecture for Layer 1 and then Layer 2 and again interconnect the layers with the hierarchical relationships. This process reveals what’s next, right? Derive the requirements for each layer. If subsystem requirements don’t yet exist , derive them from the functional architecture you just created and connect them to the functions they’re based on. Don’t forget about verification, as your requirements and architecture develop you must always think of the verification for each requirement and maintain the traceability. As the STRATA process unfolds and the model builds more depth, the missing pieces of the system design will start to become evident as holes in the relationships become visible.
Step 2: Layered Decomposition – Breaking Down the System
STRATA organizes systems into layers of increasing granularity, ensuring structured decomposition while maintaining traceability. While the framework suggests a set of default layers, the names and structure are completely customizable to fit your organization’s terminology and needs. The default layers include:
- System Context Layer (L0) – Defines the external environment and mission.
- System Layer (L1) – Captures high-level system requirements and architecture.
- Subsystem Layer (L2) – Breaks down the system into functional subsystems.
- Component Layer (L3 and beyond) – Defines individual components and their interactions.
Each layer builds upon the previous one, ensuring logical consistency while allowing for iteration between levels. How each layer builds up becomes obvious after you’ve captured the initial systems knowledge in the model. Look for the gaps: are the requirements complete? Are you missing the verification aspects of your System Context Layer (L0)? Don’t you need to ensure you can verify those system level requirements? Have the behavioral aspects of the layer been fully developed?
Model to the level that makes sense for your needs. This is sometimes easier said than done and often it’s asked, “when do I stop modeling”, There honestly is no simple answer to this question, but if you’re modeling for its own sake, then you’ve probably gone too far. Modeling a layer should be focused on the output of that layer. If you need to create a subsystem specification, then model completely through the subsystem layer, and you’ll find the data needed to develop that specification. When you can answer the need, you’ve completed a layer. The goal of each layer is to produce the necessary products, which can be a specification, interface control document, V&V plan, etc…. or all these products.
Step 3: Cross-Pillar Alignment – Ensuring Consistency
STRATA integrates four core perspectives:
- Requirements – Capturing stakeholder needs and constraints.
- Functional Architecture & Behavior – Defining system operations and interactions.
- Physical Architecture – Structuring components and their relationships.
- Verification & Validation (V&V) – Ensuring correctness through testing and analysis.
Changes in one pillar impact the others, requiring continuous alignment. STRATA ensures that modifications at lower levels propagate correctly to higher levels, preventing inconsistencies. As you model a layer, a key aspect of STRATA is to ensure you align the aspect of the layer you’re working on with the layer above and below it. Understanding the relationship between each aspect of the model and its parent and child layer is critical to understanding the full perspective. Confirming that each layer is decomposing the parent layer and adding value and understanding can only be done by understanding the relationship to the parent level.
Too often concepts and model aspects are decomposed with little thought as to why they were decomposed as shown in the model in the first place. Keeping the traceability to the parent aspect of the aspect you’re working on ensures that the decomposition is correct and being done to provide a deeper understanding of your system and not just done arbitrarily.
Understanding the impact of a change is crucial in systems engineering, ensuring the traceability between levels gives the systems engineer clear insight into the impact of changes. When something changes to the level you’re currently working on, you can instantly assess its impact on the lower level and ensure the change is being made in accordance with the higher level. This sounds amazingly like common sense, but in practice it is surprisingly often forgotten as engineers tend to focus on a single physical layer at a time and tend to lose sight of the reason for that physical architecture.
Step 4: Iterative Refinement – Managing Complexity
Unlike traditional waterfall methodologies, STRATA recognizes that system development is iterative. Engineers must:
- Refine requirements as new insights emerge.
- Adjust architecture based on feasibility analysis.
- Validate assumptions through simulations and testing.
- Ensure traceability between layers to maintain system integrity.
STRATA’s structured approach ensures that changes are systematically managed, preventing costly rework and misalignment. Complete the layers to the best of your knowledge at the time. The model, and each layer in it, will grow and change as you and your team’s knowledge grows as well. Don’t be afraid to re-work portions of a layer as new knowledge and understanding becomes available, use relationships to other aspects of the system to help inform the updates and complete each layer while maintaining the traceability up and down a layer.
Step 5: Using GENESYS to Support STRATA Implementation
Vitech’s GENESYS tool plays a crucial role in STRATA implementation by:
- Managing relationships between system elements.
- Generating diagrams to visualize dependencies.
- Supporting simulations to validate system behavior.
- Providing consistency checks to ensure model integrity.
By integrating STRATA with GENESYS, engineers can streamline system development, ensuring that all elements remain aligned throughout the process. GENESYS allows the engineer to maintain the holistic view of the system even in the face of the many layers of decomposition through the natural and completely integrated way it puts relationships and entities first. Models are easy to navigate through the relations, as all entity’s relations to any other entity are clearly shown to the engineer.
Diagrams are then constructed directly from the underlying model. Conversely, when the engineer creates a diagram, the model is updated accordingly. This language-first centricity is core to how GENESYS works, and provides unparalleled insight into your systems, while ensuring consistency. Models cannot become inconsistent in GENESYS; the language-first approach guarantees it.
Conclusion
Implementing STRATA MBSE requires a structured, iterative approach that balances layered decomposition, cross-pillar alignment, and continuous refinement. By leveraging CSDL, GENESYS, and strategic layering, engineers can manage complexity effectively, ensuring that systems are developed efficiently and consistently.
In our next blog, we’ll explore real-world applications of STRATA, showcasing how organizations use this methodology to optimize system design and integration.
