Institute of Complex Systems (iCoSys)
[Adapted from "Ultra-Large-Scale Systems: The Software Challenge of the Future", Carnegie Mellon University, June 2006 ]
The basic premise underlying our approach is that beyond a certain threshold of complexity, a traditional central engineering approach is no longer adequate and can no longer be the primary means by which complex systems can become a reality.
- Power grids, water distribution networks, transport networks are designed, but cities are not - even if their form is regulated by both the natural constraints and imposed constraints.
- Companies are designed, but the overall structure of the economy is not-even if it is regulated.
- Ecosystems exhibit a very complex organization which is not due to engineering.
- The protocols on which the Internet is based were designed, but the Web as a whole was not designed but its shape is limited by natural and artificial constraints.
Complex systems require a radical change of perspective, the satisfaction of needs through traditional rational means and top-down engineering, has to be replaced by control (orchestration) of complex distributed systems.
In other words, a system is said "complex" when it consists of a large number of interacting subsystems for which it is difficult to establish simple rules explaining its behavior or evolution, one also speaks of "Systems of Systems".