Fundamentals of Holonic Systems and Their Implications for Self-Adaptive and Self-Organizing Systems

Abstract

Holonic systems originate from Simon's research on the Sciences of the Artificial [4]. In a demanding and dynamic environment, all non-trivial systems must posses a holonic (pyramidal) structure to adapt swiftly while the information processing capacity is bounded. This paper discusses the implications of the manner in which holonic systems emerge, adapt and survive on the design of self-adaptive and self-organizing systems. First, self-adaptive and self-organizing holons must be resilient against the dynamics of their surroundings, deliver services within the setting of their surrounding super-holon(s), and shield neighboring holons from dynamics that those neighbors cannot digest. This paper discusses a holonic design illustrating this. Second, a holonic system design must balance self-*elements with more conventional elements. Indeed, the more efficient conventional elements increase the adaptation speed. This paper discusses how self adaptive and self-organizing holons can be combined with other holons without forfeiting the qualities of these self-* elements (operating range, low-and-late commitment).