Effects of topological structure and destination selection strategies on agent dynamics in complex networks

Abstract

We analyzed agents' behavior in complex networks: Barab\'asi-Albert, Erdos-R\'enyi, and Watts-Strogatz models. We observed three scenarios: (a) Agents randomly selecting a destination among adjacent nodes. (b) Excluding the most congested adjacent node as a destination. (c) Always choosing the sparsest adjacent node as a destination. We measured the hunching rate, that is, the rate of change of agent amounts in each node per unit of time, and the imbalance of agent distribution among nodes. Our empirical study reveals that topological network structure precisely determines agent distribution when agents perform full random walks; however, their destination selections alter the agent distribution. (c) makes hunching and imbalance rates significantly higher than those in random walk scenarios when the network has a high degree. (b) increases the hunching rate while decreasing the imbalance rate when activity is low; however, both increase when activity is high. These physical characteristics exhibited periodic undulations over time.