Properties of Complex Adaptive Systems |

Properties of Complex Adaptive Systems

Agents (Stakeholders, firms, organisations, humans, animals, cells, molecules etc.).	The number of agents in a system is not fixed. They are goal-directed (their behaviour aims to maximize their individual “fitness”, “utility” or “preference). They learn and adapt.The action of one agent will trigger actions by other agents – setting in motion a chain of activity. Starting from the local to the wider system.

Non-linearity

The effects of behaviour are not proportional to the causes. When the effects are larger than the causes there is amplification (positive feedback) – small perturbations can be reinforced and intensify – like a single infection can become a pandemic. The reverse, where the effects are smaller than the causes (negative feedback), there is a dampening, where the perturbations are gradually dampened. In the case of positive feedback – interactions can be sensitive to their initial conditions – a tiny undetectable change can result in a dramatic outcome. The dynamics can be a combination of both negative and positive feedback. A system might be thriving but can suddenly collapse

Emergence

The whole has emergent properties – Water (H₂O) consists of three molecules – individually they are nothing like water. The individual agents (molecules in this case) obey new rules – they lose individual autonomy and function to minimize friction between the agents and maximise their collective fitness, preference or utility. Synergy is preferred to individual freedom – the selfish agent understands the preferred state of cooperation. The cohesive whole is more than the sum of the parts. – there are benefits and unknown outcome.

Feedback	See non-linearity above.

Self-organize

Systems spontaneously organize themselves into new structures and configurations. They adjust to cope with changes in the internal and external environment. The process is collective (parallel and distributed) – it enables the system to be robust and resilient to perturbations.It starts with only the local (immediate neighbours) interactions. Remote agents at first are independent, but because of the interconnections, changes propagate to distant sub-systems. The effect is unpredictable due to the nature of feedback.

Redundancy	The ability to withstand perturbation without damage.

Unpredictability	Pattern of behaviour that a system displays in response to perturbations.

Interconnectivity	The relationships between agents are more important than the agents themselves.

Constraints	A force (loosely defined – information, budget etc.) restricting the movement of a system.

Attractors	Phenomena that stimulates agents to gravitate towards them, like magnetic forces. The system’s trajectory is irregular and unknowable towards the vicinity of where the strange attractor appears to be at a given iteration.

Boundaries	Parameters provide boundaries that signal containment and what agents and the system can interact with.

Rules	Each agent follows local rules. They change in response to the environment – not arbitrarily.

Autonomy	Agents and the system’s ability to act differently despite identical circumstances.

Robustness	The system’s ability to function despite internal or external changes, ambiguity, or lack of information. It is able to self-repair, self-regulate etc.

Disposition	Inclination/tendency for a certain type of behaviour. The probability about how the system might change. Systems behaviour cannot be predicted.