ABSTRACT
Self-organization is increasingly being regarded as an effective approach to tackle the complexity of modern systems. The self-organization approach allows the development of systems exhibiting complex dynamics and adapting to environmental perturbations without requiring a complete knowledge of the surrounding conditions to come. Systems developed according to self-organization principles are composed by simple entities that, locally interacting with others sharing the same environment, collectively produce the target global patterns and dynamics by emergence. Many biological systems can be effectively modeled using a self-organization approach: well-known examples include food foraging in ant colonies, nest building in termites societies, comb pattern in honeybees, brood sorting in ants [Parunak, 1997; Bonabeau et al., 1999; Camazine et al., 2001]. They have inspired the development of many artificial systems, such as decentralized coordination for automated guided
vehicles [Sauter et al., 2005; Weyns et al., 2005], congestion avoidance in circuit switched telecommunication networks [Steward and Appleby, 1994], manufacturing scheduling and control for vehicle painting [Cicirello and Smith, 2004], and self-organizing peer-to-peer infrastructures [Babaoglu et al., 2002]. Furthermore, principles of self-organization are currently investigated in several research initiatives that may have industrial relevance in a near future: notable examples include Amorphous Computing [Abelson et al., 2000], Autonomic Computing [Kephart and Chess, 2003] and Bioinformatics.
