In Chapter 2, we explored system thinking and how to think about things as systems. A number of key ideas were introduced in Chapter 2, including the power of systems, form and function, entities and relationships, abstractions and emergence, and boundaries and context.

The approaches described in Chapter 2 were deliberately demonstrated on simple abstractions of systems with two or three important entities and relationships. This was done to focus the reader’s attention on the system issues rather than on the high level of complexity of systems we normally encounter.Many of the key concepts of systems thinking (such as the power of systems, form and function, entities and relationships, abstractions and emergence, and boundaries and context) can be demonstrated in simple abstractions of systems. However, most of the systems that we deal with professionally are complex. These include the artifacts we design, the systems used to build and operate them, the stakeholder systems involved, and the organizations in which we work. Hence, in this chapter we will extend the approaches of system thinking to more complex systems. In other words, we will start developing the methods of system architecting.

Complexity is inherent in system architecture, and before proceeding with a rigorous development of the analysis of architecture (Part 2) and the synthesis of architecture (Part 3), we should spend a short time dealing with the issues of complexity in systems and the approaches that have been developed to enable us to comprehend complex systems.

This chapter starts with a brief discussion of what makes systems complex. We then summarize the approaches that enable people to better understand and work with complex systems. These include decomposition and hierarchy, different types of relationships among entities (such as class/instance relationships and specialization/generalization), and specific tools to reason through complex systems, such as zigzagging, views, and projections. The chapter ends with an introduction to SysML and OPM: two tools that are often used to represent complex systems.