Sampling and active learning methods for network reliability estimation using K-terminal spanning tree

TL;DR

This paper introduces a novel sampling and active learning approach for efficient network reliability estimation, leveraging Monte Carlo sampling, K-terminal spanning trees, and a random forest classifier to improve accuracy and adaptability across network topologies.

Contribution

It proposes a new sampling method combined with an active learning framework using random forests to enhance network reliability estimation efficiency and adaptability.

Findings

The methods outperform existing techniques in accuracy and efficiency.

The RF classifier effectively predicts reliability across different network topologies.

The approach is validated on multiple network examples and practical applications.

Abstract

Network reliability analysis remains a challenge due to the increasing size and complexity of networks. This paper presents a novel sampling method and an active learning method for efficient and accurate network reliability estimation under node failure and edge failure scenarios. The proposed sampling method adopts Monte Carlo technique to sample component lifetimes and the K-terminal spanning tree algorithm to accelerate structure function computation. Unlike existing methods that compute only one structure function value per sample, our method generates multiple component state vectors and corresponding structure function values from each sample. Network reliability is estimated based on survival signatures derived from these values. A transformation technique extends this method to handle both node failure and edge failure. To enhance efficiency of proposed sampling method and achieve adaptability to network topology changes, we introduce an active learning method utilizing a random forest (RF) classifier. This classifier directly predicts structure function values, integrates network behaviors across diverse topologies, and undergoes iterative refinement to enhance predictive accuracy. Importantly, the trained RF classifier can directly predict reliability for variant networks, a capability beyond the sampling method alone. Through investigating several network examples and two practical applications, the effectiveness of both proposed methods is demonstrated.