Dynamics of Complex Systems Built as Coupled Physical, Communication and Decision Layers

Area of Science:

• Complex Systems
• Agent-Based Modeling
• Network Science

Background:

• Multilayer systems exhibit intricate interdependencies between constituent layers.
• Individual agents within these systems often pursue self-optimization goals.
• Understanding emergent behaviors in such systems is crucial for design and control.

Purpose of the Study:

• To develop a simplified model for analyzing multilayer systems with interacting layers.
• To investigate how individual agent behavior influences system-level dynamics.
• To explore the impact of system size, network topology, and communication errors on system performance.

Main Methods:

• An agent-based simulation was developed to model a physical layer (power source and resistors) and a decision layer for agents.
• Agents dynamically adjusted resistors to maximize delivered power, influenced by internal state, global perception, neighbor communication, and randomized selfishness.
• Key parameters analyzed included system size, communication network topology, communication error rates, and minimum power gain thresholds.

Main Results:

• Emergent wave-like macro-level behavior was observed, but only within a specific range of system sizes.
• The ratio of cooperative to non-cooperative agents was sensitive to the minimum power gain threshold, with lower thresholds reducing cooperation.
• Network topology significantly affected power utilization and fairness, while communication errors unexpectedly promoted higher levels of cooperation.

Conclusions:

• System size is a critical factor for the emergence of coordinated macro-level behaviors in these multilayer systems.
• Agent decision thresholds and communication network characteristics play vital roles in system efficiency and fairness.
• Introducing controlled communication errors can be a strategy to enhance cooperation within the system.