The Evolutionary Design Model (EDM) for the design of complex engineered systems : Masdar City as a case study

Abstract

This thesis develops a framework for constructing an Evolutionary Design Model (EDM) that would enhance the design of complex systems through an efficient process. The framework proposed is generic and suggests a group of systematic methodologies that eventually lead to a fully realized and integrated design model. Within this model, complexities of the design are handled and the uncertainties of the design evolution are managed. Using the framework, vast design spaces can be searched while solutions are intelligently modified, their performance evaluated, and their results aggregated into a compatible set for design decisions. The EDM is composed of several design states as well as design evolving processes. A design state describes a design at a particular point in time and maps the system's object to the system's requirements and identifies its relation to the context in which the system will operate. A design evolving process involves many sub-processes which include formulation, decomposition, modeling, and integration. These sub-processes are not always carried out in a sequential manner, but rather a continuous move back and forth to previous and subsequent stages is expected. The resulting design model is described as an evolutionary model that moves a system's design from simple abstract states to more complex and detailed states throughout its evolution.

(cont.) The framework utilizes system modeling methodologies that include both logical and mathematical modeling methods. The type of model used within the EDM's evolving processes is highly dependent on and driven by design needs of each process. As the design progresses a shift from logical models to mathematical models occurs within the EDM. Finally, a partial EDM is implemented within the context of a computational design system for Masdar city to demonstrate the application of the proposed framework.