Modelling cascading failures in networks with the harmonic closeness

Area of Science:

• Network science
• Complex systems analysis
• Systems engineering

Background:

• Traditional load definition in cascading failure studies relies on node degree or betweenness.
• Existing methods may not fully capture the intricate dynamics of network load distribution.

Purpose of the Study:

• To propose a novel approach for initial load distribution in networks.
• To evaluate the effectiveness of this new method in preventing cascading failures.
• To compare the proposed method against existing approaches.

Main Methods:

• Developed a new load distribution model incorporating harmonic closeness and neighboring node impact.
• Conducted simulations on scale-free (SF), small-world (SW), and Erdos-Renyi (ER) networks.
• Analyzed the influence of adjustable parameters (θ, δ) and the proportion of attacked nodes (f).

Main Results:

• Identified a negative correlation between optimal parameters θ and δ in SF, SW, and ER networks.
• Demonstrated that removing low-load nodes can sometimes increase cascading failures.
• The proposed harmonic closeness method shows superior performance in SF, SW, and ER networks compared to other methods, especially for certain parameter ranges.
• The method enhances network robustness across different average degrees.

Conclusions:

• The novel harmonic closeness-based load distribution method is effective in mitigating cascading failures.
• This approach offers improved robustness and performance across diverse network topologies, including real-world networks.
• The findings suggest a new direction for designing resilient network infrastructures.