Yuxian Du

A new closeness centrality measure via effective distance in complex networks

Closeness centrality (CC) measure, as a well-known global measure, is widely applied in many complex networks. However, the classical CC presents many problems for flow networks since these networks are directed and weighted. To address these issues, we propose an effective distance based closeness centrality (EDCC), which uses effective distance to replace conventional geographic distance and binary distance obtained by Dijkstras shortest path algorithm. The proposed EDCC considers not only the global structure of the network but also the local information of nodes. And it can be well applied in directed or undirected, weighted or unweighted networks. Susceptible-Infected model is utilized to evaluate the performance by using the spreading rate and the number of infected nodes. Numerical examples simulated on four real networks are given to show the effectiveness of the proposed EDCC.

A new method in failure mode and effects analysis based on evidential reasoning

The traditional failure mode and effects analysis (FMEA) is determined by risk priority number (RPN), which is the product of three risk factors occurrence (O), severity (S), and detection (D). One of the open issues is how to precisely determine and aggregate the risk factors. However, the traditional FMEA has been extensively criticized for various reasons. In this paper, a new method in fuzzy FMEA is proposed using evidential reasoning (ER) and the technique for order preference by similarity to ideal solution (TOPSIS). The ER approach is used to express the experts’ assessment information which may be imprecise and uncertain. Considering the experts’ weights, we construct the group assessment. Weighted average method is then utilized to transform the group assessment value into crisp value. TOPSIS is applied to aggregate the risk factors which are taken to account as the multi-attribute, and used to rank the risk priority. By making full use of attribute information, TOPSIS provides a cardinal ranking of alternatives, and does not require the attribute preferences are independent. A numerical example shows that the proposed method is efficient to its applications.

Fault tree analysis based on TOPSIS and triangular fuzzy number

Fault tree analysis (FTA) is widely used in the failure probability evaluation of a system. The conventional failure probabilities of basic events are treated as crisp values. However, in many real applications, it is often difficult to evaluate failure probabilities of basic events from past occurrences. In order to address this issue, a new FTA method based on the technique for order preference by similarity to an ideal solution and the triangular fuzzy number is presented. Compared with the existing method, our proposed method is more efficient with less complexity.