Requirements analysis in tropos: a self-referencing example

• Authors: Paolo Bresciani;Fabrizio Sannicolò
• Venue: NODe'02 Proceedings of the NODe 2002 agent-related conference on Agent technologies, infrastructures, tools, and applications for E-services
• Year: 2002

As result of the process of evolution driven by the law of synergy, emergence endows the dynamics of composite systems with properties unidentifiable in their individual parts. The phenomenon of emergence involves: -- self-organization of the dynamical systems such that the synergetic effects can occur; -- interaction with other systems from which the synergetic properties can evolve in a new context. Multi-agent Systems enable cloning of real-life systems into autonomous software entities with a 'life' of their own in the dynamic information environment offered by today's Cyberspace. After introducing the concept of Holonic Enterprise (HE) as a paradigm for the networked world that enables virtual representation of real-life organizations as multi-agent systems I will present a fuzzy-evolutionary approach which mimics emergence in Cyberspace as follows: -- it induces self-organizing properties by minimizing the entropy measuring the information spread across the virtual system/organization such that equilibrium is reached in an optimal interaction between the system's parts to reach the system's objectives most efficiently; -- it enables system's evolution into a better one by enabling interaction with external systems found via genetic search strategies (mimicking mating with most fit partners in natural evolution) such that the new system's optimal organizational structure (reached by minimizing the entropy) is better then the one before evolution. The holonic enterprise paradigm provides a framework for information and resource management in global virtual organizations by modeling enterprise entities as software agents linked through the internet. Applying the proposed fuzzy-evolutionary approach to the virtual societies 'living' on the dynamic Web endows them with behavioral properties characteristic to natural systems. In this parallel universe of information, enterprises enabled with the proposed emergence mechanism can evolve towards better and better structures while at the same time self-organizing their resources to optimally accomplish the desired objectives.