The Goals

The goals of complex systems research are to understand

• Understand the development and mechanisms of patterns of behavior and their use in engineering

• Understand the way to deal with complex problems (engineering, management, economic, sociopolitical) using strategies that relate the complexity of the challenge to the complexity of the system that must respond to them

• Understand the unifying principles of organization, particularly for systems that deal with large amounts of information (physical, biological, social, and engineered)

• Understand the interplay of behaviors at multiple scales and between the system and its environment

• Understand what is universal and what is not, when averaging applies and when it does not, what can be known and what cannot, what are the classes of universal behavior and the boundaries between them, and what are the relevant parameters for describing or affecting system behaviors

• Develop the ability to capture and represent specific systems rather than just accumulate data about them: (in this context) to describe relationships, know key behaviors, recognize relevance of properties to function, and simulate dynamics and response.

• Achieve a major educational shift toward unified understanding of systems and patterns of system behavior.

The traditional approach of science of taking things apart and assigning the properties of the system to its parts has been quite successful, but the limits of this approach have become apparent in recent years. When properties of a system result from dependencies and relationships but we assign them to their parts, major obstacles arise to understanding and control. Once the error of assignment is recognized, some of the obstacles can be overcome quickly, while others become subjects of substantive inquiry. Many scientists think that the parts are universal but the way parts work together is specific to each system. However, it has become increasingly clear that how parts work together can also be studied in general, and by doing so, we gain insight into every kind of system that exists, including physical systems like the weather as well as biological, social, and engineered systems.

Understanding complex systems does not mean that we can predict their behavior exactly; it is not just about massive databases or massive simulations, even though these are important tools of research in complex systems. The main role of research in the study of complex systems is that of recognizing what we can and cannot say about complex systems given a certain level (or scale) of description and knowing how we can generalize across diverse types of complex systems. It is just as important to know what we can know, as to know. Thus the concept of deterministic chaos appears to be a contradiction in terms: how can a deterministic system also be chaotic? It is possible because there is a rate at which the system behavior becomes dependent on finer and finer details (Cvitanovic 1989; Strogatz 1994; Ott 1993). Thus, how well we know a system at a particular time determines how well we can predict its behavior over time. Understanding complexity is neither about prediction or lack of predictability, but rather a quantitative knowledge of how well we can predict, and only within this constraint, what the prediction is.

0 0

Post a comment