Goals / Ideas of the Project Quantitative Emergence (QE)

The project Quantitative Emergence focuses on the design of concepts and tools for the implementation of an architecture needed for the realisation of self-organising technical systems, which are at the same time reliable, robust, and adaptive. As an important prerequisite for designing such systems we have to understand the effects of emergent global behaviour in networks of intelligent autonomous units. Hence, we have to quantify emergent behaviour and we need tools to prevent unwanted behaviour and to encourage or enforce desired positive effects. The architecture, we develop in this project, will consist of a network of autonomous units (called "system under observation and control" (SuOC)), augmented by a combination of one or more observers and controllers, potentially structured into several layers. We develop an appropriate methodology for observing the (global) behaviour of the system and for quantifying and evaluating effects of emergence. Furthermore, we generate adequate responses of the controller to the results of the observer in order to enable a controlled emergent behaviour within the restrictions set by some external unit (the environment). This requires an exploration of various potential ways of influencing the behaviour of a self-organising production system. For the initial 2 years of this project we abstract from realistic technical applications and use rather simple artificial production systems exhibiting some interesting properties. Only later, we intend to combine the new concepts and tools for the observer/controller architecture with more complex scenarios like the organic production system "Organic Traffic Control", which is being developed in the corresponding project OTC.