Development of cascade: A multidisciplinary design test simulator

Industry is constantly pursuing faster and cheaper means for designing and manufacturing high quality goods and services. This becomes ever more difficult as the complexity of the product increases. Much of the method development in the field of Multidisciplinary Design Optimization (MDO) attempts to simplify the design of a large, complex system, by dividing the system into a series of smaller, simpler, and coupled subsystems. These methodologies require extensive testing, either on analytical representations of real-life systems, or on the actual system itself, prior to implementation in order to verify their feasiblity and robustness. Due to the complexity of these real-life systems, however, it is not practical to perform methodology feasibility studies on the analytical and numerical representations of the true system. Hence, some representative yet efficient means of determining the feasibility and robustness of MDO methods is crucial. This paper describes the design of a test simulator, CASCADE (Complex Application Simulator for the Creation of Analytical Design Equations), that is capable of randomly generating and then converging a complex system of analytical equations, of user-specified size. CASCADE-generated systems can be used to test sequencing and system reduction strategies, convergence strategies, and MDO methods, among others. Further, CASCADE has been designed to operate in a distributed computing environment (using Parallel Virtual Machine), as the field of MDO itself is perfectly suited for such a setting. This paper describes the simulator and many of its applications.