Multi-Agent Modeling of Dynamical Systems: A Self-organized, Emergent, Homeostatic and Autopoietic Approach

TL;DR

This thesis develops a multi-agent modeling approach for dynamical systems that captures properties like emergence, self-organization, and homeostasis, validated on various systems, and offers an accessible methodology for scientific analysis.

Contribution

It introduces a novel, comprehensive methodological framework for modeling dynamical systems using agents, emphasizing properties like autopoiesis and complexity with simple calculations.

Findings

Validated on boolean networks and cellular automata

Defined metrics for system complexity and properties

Demonstrated applicability across scientific fields

Abstract

This thesis presents the theoretical, conceptual and methodological aspects that support the modeling of dynamical systems (DS) by using several agents. The modeling approach permits the assessment of properties representing order, change, equilibrium, adaptability, and autonomy, in DS. The modeling processes were supported by a conceptual corpus regarding systems dynamics, multi-agent systems, graph theory, and, particularly, the information theory. Besides to the specification of the dynamical systems as a computational network of agents, metrics that allow characterizing and assessing the inherent complexity of such systems were defined. As a result, properties associated with emergence, self-organization, complexity, homeostasis and autopoiesis were defined, formalized and measured. The validation of the underlying DS model was carried out on discrete systems (boolean networks and cellular automata) and ecological systems. The central contribution of this thesis was the development of a methodological approach for DS modeling. This approach includes a larger set of properties than in traditional studies, what allows us to deepen in questioning essential issues associated with the DS field. All this was achieved from a simple base of calculation and interpretation, which does not require advanced mathematical knowledge, and facilitates their application in different fields of science.