The CHESS Philosophy
CHESS adopts a specific methodology for the design of the system, which is briefly summarized here. Further details can be found on the CHESS Website and especially in the CHESS Toolset User Guide.
The CHESS methodology is based on three pillars:
- Component-Based Development: The system is composed of reusable components that communicate only through their compatible ports.
- Correctness-by-Construction: The system design should be correct from the beginning. This is achieved by restricting the actions permitted to the user/modeler.
- Separation of Concerns: Different concerns in the design of the sytem whould be treated separately. This is achieved by the separation of the design space in different design views.
Modeling is performed using the CHESS Modeling Language (CHESS ML), a custom language based on elements from UML, SysML, and MARTE, as well as specific stereotypes defined within the project.
The customized version of the Papyrus editor enforces the separation of different design views. In each of them only a certain set of elements are avaialble in the tool palette. The selection of principal design views is performed using the view-aware model explorer. The current view is shown by a colored indicator in the toolbar.
In addition, domain-specific (sub-)views are activated by specific buttons in the toolbar. In the image below, the Extra-Functional View can be activated as part of the Component View.
Currently, the following views are available in PolarSys CHESS. We highlight in bold those that are relevant for the CHESS-SBA workflow.
- Requirement View
- System View
-
Component View
- Extra-Functional View (toolbar)
-
Deployment View
- Dependability View (toolbar)
- Railway View (toolbar)
- Analysis View
- Dependability Analysis View
- Real-Time Analysis View
- PSM View
The purpose of each view is detailed later when describing the CHESS-SBA modeling extensions.
You can now proceed to learn how the system is modeled using CHESS ML.