03 September 2014

Principles of Systems Thinking - SEBOK (21)

This topic forms part of the Systems Thinking Knowledge Area (KA). It identifies systems principles as part of the basic ideas of systems thinking. Some additional concepts more directly associated with engineered systems are described, and a summary of system principles associated with the concepts already defined is provided.

Systems Principles, Laws, and Heuristics

A principle is a general rule of conduct or behavior (Lawson and Martin 2008). It can also be defined as a basic generalization that is accepted as true and that can be used as a basis for reasoning or conduct (WordWeb 2012c). Thus, systems principles can be used as a basis for reasoning about systems thinking or associated conduct (systems approaches).

Summary of Systems Principles

Regularity (glossary): Systems science should find and capture regularities in systems, because those regularities promote systems understanding and facilitate systems practice. (Bertalanffy 1968)

Holism (glossary): A system should be considered as a single entity, a whole, not just as a set of parts. (Ackoff 1979; Klir 2001)

Interaction The properties, capabilities, and behavior of a system are derived from its parts, from interactions between those parts, and from interactions with other systems. (Hitchins 2009 p. 60)

Relations A system is characterized by its relations: the interconnections between the elements. Feedback is a type of relation. The set of relations defines the network of the system. (Odum 1994)

Boundary (glossary): A boundary or membrane separates the system from the external world. It serves to concentrate interactions inside the system while allowing exchange with external systems. (Hoagland, Dodson, and Mauck 2001)

Synthesis (glossary): Systems can be created by choosing (conceiving, designing, selecting) the right parts, bringing them together to interact in the right way, and in orchestrating those interactions to create requisite properties of the whole, such that it performs with optimum effectiveness in its operational environment, so solving the problem that prompted its creation” (Hitchins 2008: 120).

Abstraction (glossary): A focus on essential characteristics is important in problem solving because it allows problem solvers to ignore the nonessential, thus simplifying the problem. (Sci-Tech Encyclopedia 2009; SearchCIO 2012; Pearce 2012)

Separation of Concerns

A larger problem is more effectively solved when decomposed into a set of smaller problems or concerns. (Erl 2012; Greer 2008)

View (glossary) Multiple views, each based on a system aspect or concern, are essential to understand a complex system or problem situation. One critical view is how concern relates to properties of the whole. (Edson 2008; Hybertson 2009)

Modularity (glossary): Unrelated parts of the system should be separated, and related parts of the system should be grouped together. (Griswold 1995; Wikipedia 2012a)

Encapsulation Hide internal parts and their interactions from the external environment. (Klerer 1993; IEEE 1990)

Similarity/Difference

Both the similarities and differences in systems should be recognized and accepted for what they are. (Bertalanffy 1975 p. 75; Hybertson 2009). Avoid forcing one size fits all, and avoid treating everything as entirely unique.

Dualism (glossary): Recognize dualities and consider how they are, or can be, harmonized in the context of a larger whole (Hybertson 2009)

Leverage (glossary): Achieve maximum leverage (Hybertson 2009). Because of the power versus generality tradeoff, leverage can be achieved by a complete solution (power) for a narrow class of problems, or by a partial solution for a broad class of problems (generality).

Change Change is necessary for growth and adaptation, and should be accepted and planned for as part of the natural order of things rather than something to be ignored, avoided, or prohibited (Bertalanffy 1968; Hybertson 2009).

Stability/ Change Things change at different rates, and entities or concepts at the stable end of the spectrum can and should be used to provide a guiding context for rapidly changing entities at the volatile end of the spectrum (Hybertson 2009). The study of complex adaptive systems can give guidance to system behavior and design in changing environments (Holland 1992).