Model-based Systems Engineering (MBSE) works with different model interpretations, such as simulation models and architectural models. Both model types are essential to building successful, complex systems comprising hardware, software, electrical, and mechanical elements. Yet, to date, these models have largely inhabited very separate worlds. Fortunately, these perspectives are (slowly) moving towards each other in the form of specification models to connect the worlds of architecture and simulation, concentrating on the advantages of each of the modeling concepts. As a result, engineering organizations are able to create a holistic view of a system design.
Coming Up with a Master Plan for Today’s Complex Systems
Models are critical to solving system engineering problems efficiently. For many decades, system engineers were the masters of many and connected documents (i.e., requirements, specification, interface, test, etc.). However, it was challenging to confidently confirm the design fulfilled all requirements using these text-based documents, especially since requirements tend to change continuously. Yet it’s essential to validate and verify that all requirements (even newly introduced ones) match the system design.
Fast-forward to today’s increasingly complex systems, and that challenge gets compounded. In fact, it’s fair to say that it’s simply not possible to understand how all concepts and artifacts work together when relying on documents. Many of the components and functions that were once quite separate are now tied to others. Moreover, integration occurs when the system is implemented. With sub-functionalities distributed throughout a system, a single change can impact many others on all levels.
Everyone involved in bringing that system to market needs access to a common model that helps trace all these relationships. However, traceability is not the only important objective. To stay within time and budget constraints, engineering organizations cannot afford to repeat efforts; they must reuse wherever possible. That requires a master plan of all parts, concepts and artifacts.
Bridging the Architecture and Simulation Models
Such a master plan needs to take into account the unique characteristics and requirements of both simulation models and architectural models.
Simulation models are usually built up on a specific construction paradigm, namely a functional decomposition. This model – which builds up one system to fulfill all requirements –works when assessing a system’s functions. However, it’s challenging to easily reuse a specific part within this model because everything is tied together.
On the other hand, object orientation rather than functional decomposition is used for software architecture in SysML and MBSE models. This helps distinguish between structural or behavioral types and usage of these types. If an engineer changes one type of the system, all usages will also be changed. At the same time, it’s fairly easy to reuse a type, especially as generalizations of types exist as well. Consider the many types of screws in existence. These are complemented by screwdrivers, which all include a handle and tip that accommodates the given screw.
The way to bridge these two important models is using a specification model for the overall architecture built on black boxes, and building in the simulation models as white boxes.
Providing Much-Needed Perspective
Specification models help engineers and others focus on a specific perspective while all relationships are maintained consistently in the underlying, possibly, complicated model. Essentially, they build the base structure of a whole system (or a system of systems). This is critical in bringing together the worlds of design engineering and simulation.
Think of system views or perspectives enabled by specification models as akin to people working on separate islands on a related project. A holistic view of all the people and their ways of working enables a better understanding of how everything fits together.
As the master plan, a specification model enables engineers to think about how the system perspectives fit together and trace from one abstraction layer to the next or to another. In more complex, safety-critical systems, engineering organizations must comply with many regulations and norms. To ensure compliance, they must be able to look downstream at where a requirement was implemented. Upstream they must be able to determine who dictated the requirement. The only way to achieve this is using a specification model.
In summary: Always use the right model: simulation models for complex behavior and specification models for architecture. If questions come up how to bridge these perspectives, vendors like PTC can help answer these.