Digital Ecosystems: Self-Organised Complexity of Evolving Agent Populations

A primary motivation for our research in Digital Ecosystems is the desire to exploit the self-organising properties of biological ecosystems. Ecosystems are thought to be robust, scalable architectures that can automatically solve complex, dynamic problems. Self-organisation is perhaps one of the most desirable features in the systems that we design, and it is important for us to be able to measure such self-organising behaviour. We investigate the self-organising aspects of Digital Ecosystems, created through combining Service-Oriented Architecture, distributed evolutionary computing and Multi-Agent Systems, aiming to determine a macroscopic variable to characterise the self-organised complexity of the evolving agent populations within our Digital Ecosystem. We study a definition for the self-organised 'complexity', grounded in the biological sciences, called Physical Complexity. It is based on statistical physics, automata theory, and information theory, providing a measure of the quantity of information in an organism's genome, by calculating the entropy in the population to determine the randomness in the genome. Physical Complexity is currently defined for an ensemble of sequences of the same length, and so we investigate an extension to include ensembles of variable length sequences. We then investigate the self-organising behaviour of clustering within evolving agent populations, with the creation of an efficiency measure based on our extended Physical Complexity. Overall an insight has been achieved into where and how self-organisation occurs in our Digital Ecosystem, and how it can be quantified.