Rapid Identification and Characterization of System

New technologies enable us to build systems with greater complexity. To reduce risks imposed by the increasing complexity, organizations have to rely on seamless integrations and/or collaborations to build up their enterprises or improve their existing operations. Therefore, more and more companies that traditionally thought of themselves as system developers are beginning to realize that in fact they are "only" subsystem developers. They develop subsystems, which can be either products or services, to integrate into customer's enterprise systems. To ensure adding value to customers, a "system developer" needs to understand the systems and gain a thorough understanding of system requirements for the subsystems to be designed. The main issue these "system developers" face is not being able to access information about customers' enterprise system architectures. Without such information, it is very difficult to accurately identify value proposition to customers. Since there is no scientific approach to gain such information to date, a system architect has to rely on his or her experience or intuition to make important design decisions. The thesis proposes a framework, consisting of system context analysis, system modeling and analysis, and system design and decomposition, which allows a system architect to design a class of complex systems, known as systems of systems. The term systems of systems describes the systems in which each subsystem can usually function as a standalone system. A generic oil supply chain is used throughout the thesis as an example of a system of systems. The thesis also describes the processes of developing a System Dynamics model to capture the structures and decision processes of a generic oil supply chain. It explains how a system architect can apply a scientific approach to explore the abstract oil supply chain model by conducting model testing and policy analysis. Further design confidence is developed by evaluating design options with policy analysis. More importantly, this thesis demonstrates how a better design option could be discovered during policy analysis.