Complex problems may require that many
data items need to be considered (many inputs)
Individual data items in a complex
problem domain may contribute differently to multiple parts of the same problem
domain (different functional relationships)
Data item relationships in complex
problem domains may be non-linear (non-linearity)
Solutions to complex problems may
require that many variables be simultaneously controlled (many
outputs)
Data items in complex problem domains
need to be understood both for the potential importance and for their
instantaneous value (or instantaneous importance) (time value of
information)
Complex problems may be dynamic or have
dynamic components (dynamic situations)
Solutions to complex problems may
require suboptimal solutions to component options (balancing
options)
Complex problems are often composed of a
number of component problems that must be considered before responding to
higher level problems (multi-dimensional problems)
Complex problems are often identified
with conflicting goals (conflict at same level)
Components of complex problem domains
often have goals that compete with higher level goals (conflict across
hierarchy)
Complex problems may have time and
distance involvement (temporal impact)
Complex problems may require the
selection of the best 'option' from a set of options, each of which may include
the control of analog (relative) outputs (analog control)
Solutions to complex problems may
include the generation of events (event triggers)
Events initiated as a solution to
complex problems can be relative (relative event triggers)
Components of complex problem domains
may or may not be well understood (by human domain experts)
Complex problems may contain components
(data items) that possess some level of randomness (chaos theory)
Another view of complex systems is that
they are 'self-organizing' (self-organizing)
