Some facts about Complexity from Ontonix:

The amount of fitness of a system is proportional to its complexity – higher complexity implies higher fitness

The amount of functionality of a system is proportional to complexity – more complex systems can perform

more functions

Each system can only reach a specific maximum value of complexity

Close to the upper limit the system is fragile – it is unwise to operate close to this limit

High complexity = difficulty in management – highly complex systems are able to perform more functions

but at a price: they are not easy to manage

When a system is very complex and becomes difficult to manage, it is necessary to restructure it, add new

structure or to remove excess entropy.

More components don’t necessarily imply more complexity – systems with few components can be more

complex than systems with many components.

When presented with two equivalent options, for example in terms of performance, risk or profit, select the

one with the lower complexity – it will be easier to manage.

Spasms or dramatic changes in dynamical systems are always accompanied by sudden changes in

complexity.

In nature, systems tend toward states of higher complexity, but only until they reach the corresponding

maximum. This poses limits to growth and evolution.

Systems with high complexity can behave in a multitude of ways (modes).

Systems with high complexity are more difficult to manage and control because of the need to compromise

A system with a given complexity will be more difficult to manage if it is made to operate in a more uncertain

environment.

"High complexity is incompatible with high precision": this is known as L. Zadeh’s Principle of Incompatibility.

In essence, you can’t make precise statements about a highly complex system.

A fundamental characteristic of highly complex systems: they are robust yet fragile!

Past – Prevailing[conventional – Risk Management – correlation – “Failsafe” – protection]

Current – Future[complexity – Resilience Management

– causality – “safe to fail” – prevention]

randomness

Alea

tory

Unc

erta

inty

Unf

ores

eeab

le

Irreducible(unknowable)

Use what if analyses and scenario planning based upon (endogenous) system data:

available (exogenous) statistical risk data, “line of business” claims experience and Actuarial

analyses to “model” resilience

plausible

Epis

tem

icU

ncer

tain

ty

Unf

orse

en

Partiallyor

FullyReducible

Monitor system complexity to assess what/where/how much redundancy is required to avoid fragility and maintain system resilience

Monitoring can be conducted in real-time

possibleBusiness is viewed as a dynamic complex system consisting of multiple, interdependent,

processes, sub-systems and networks

ALL, available, current “system” data processed to map & measure number, structure and

effectiveness of inter-connections

Determine state of health: -current, max. and min. sustainable complexity levels

& current resilience[if req’d. incl. ecosystem and networks]

Measurement of complexity and resilience extends the “risk horizon” into uncertainty

Identifying sources of emergent risk/opportunity; insight into causal relationships; crisis anticipation or

anticipatory awareness; a basis for loss avoidance and a culture of prevention

Supplements current analyses and facilitates earlier intervention incl. conventional RM tools & techniques to mitigate and manage identified

exposures within required tolerances

Patterns – Knowledge – Wisdom

Risk

Seen

(c

orre

latio

n) Probability based upon statistical analyses of risk data

probable

Com

plex

ity(E

pist

emic

Unc

erta

inty

)

Uns

een

(cau

salit

y)

Datainformation structure

possible

Unknown

Unknowns

Known

Knowns

Knowable &

Reducible Uncertainty

Known

Unknowns

Unknown

Knowns