Methods of managing a self-organizing network based on agent-based modeling with fuzzy logic

DOI:10.31673/2412-9070.2025.048452

Abstract

The article presents a comprehensive study of traffic flow management methods in urban envi-ronments using agent-based models and fuzzy logic. As urban traffic intensity increases each year, this leads to issues such as congestion, elevated air pollution, reduced traffic speed, and increased stress among road users. Traditional traffic management methods often fail to account for the complexity and dynamic nature of modern transport systems, where diverse agents such as cars, pedestrians, and public transport interact. Agent-based modeling allows for the creation of more flexible and adaptive models where each road user (agent) can make autonomous decisions based on local conditions.
The article emphasizes the importance of using fuzzy logic to process incom plete or vague information, which frequently occurs in real urban conditions. Fuzzy logic enables the modeling of complex traffic behaviors and improves decision-making processes in cases where traffic parameters change unpredictably. For example, fuzzy rules are applied to traffic light control, adjusting signal duration according to current traffic intensity.
Simulations conducted using the proposed model demonstrated that integrating agent-based approaches with fuzzy logic can significantly reduce congestion, shorten delay times, and improve the throughput of the transportation system under various traffic scenarios. The model was tested across multiple levels of network load, from low to high, and in all cases, it showed high efficiency. The proposed approach not only aids in analyzing current conditions but also provides opportunities for future infrastructure planning and forecasting.
The application of this model could contribute to more efficient use of road infrastructure, improve urban travel conditions, and reduce the negative environmental impacts through decreased congestion and air pollution.

Keywords: agent-based model; fuzzy logic; information networks; information systems; computer modeling; optimization; control system.