Enabling modular design platforms for complex systems

Abstract

In recent times, an emerging trend in several industries that have adopted Model-Based Design has been holistic product platforms where a single systems design is reused and customized to meet diverse customer requirements such as application, cost, and operational considerations. Many of these dynamic changes in nature have required system design component variations referred to as "variants" on top of a fixed master design. One approach to realize this is to create copies of the original design for each variant combination. Additionally, this requires a sophisticated traceability mechanism to propagate any changes in the design to the various implementations. An alternative approach is to design a modular architecture that can reference all the product variations within a single file. Different implementations can then be realized by selecting different system components through a scripting language. This approach promotes design reuse and provides a powerful mechanism to implement traceability. However, such a paradigm requires core tool functionality similar to those available in various UML/SysML implementations before being applied to a systems development process. In this paper, we introduce variant semantics for complex systems design for use within the Simulink modeling environment. We discuss their attributes which can be parametric or structural that can be used throughout the development process. In addition to improving the efficiency and development of product variations, variants present a variety of uses in the context of systems engineering workflows. For example, design exploration, where several alternatives exist for a component, can now be managed efficiently to simulate every design possibility in a combinatorial fashion for a given test suite. For large-scale problems, these simulations could be distributed to a high performance computing cluster for overall speedup through a scripting methodology. Design elaboration and integration is a challenging activity that can also be improved through the use of variants, where low fidelity components are replaced by more specialized one's going from mathematical equations to physical or software elements. Since the order in which these components are integrated influence design quality and subsequent iterations, it is possible to carry out several separate integrations that increase confidence. Since there are a number of ways to modularize a design, we also outline a set of best practices for partitioning the design variations for scalability and maintainability. Using Simulink-based examples, we illustrate the above scenarios and outline strategies on how organizations can leverage these possibilities to reuse while enhancing their existing knowledge to meet system design challenges of the future. © The MathWorks, Inc. 2011.