Software Engineering Institute

Model-based methods have been used for centuries to reduce risk in system designs and improve design quality. In the context of the embedded computing resources of Cyber-Physical Systems (CPS), designs are growing more and more complex. The software that is incorporated within the system becomes a larger part of the overall technical solution. The number of physical parameters that are monitored and controlled by the system contribute to this complexity, which results in potentially hard-to-predict behavior: It is very difficult to understand the suitability of a proposed system solution without a deep understanding of how its computing resources are going to be used (as part of the overall technical solution). In this webcast, Fred Schenker, Jerome Hugues, and Linda Parker Gates discuss the benefits of using a model-based approach to improve the design of a CPS’ embedded computing resources. This is accomplished by (1) building virtual architectural models of the CPS’ embedded computing resources early in the system development lifecycle and (2) using these models to predict computing system constraints and component integration issues. They will discuss the cultural resistance to adopting the model-based approach, and how established justification methods, e.g., Return on Investment, are being used to stifle the adoption. Finally, some alternatives to ROI will be proposed that would be more effective justification mechanisms.

What attendees will learn:

• The benefits of using a model-based approach to the design of a CPS’ embedded computing resources
• The “status quo” response to any proposed process change (where is the ROI?)
• Problems with ROI as a justification mechanism
• Alternatives to ROI to justify the model-based approach