A Novel Approach to Ship Operational Risk Analysis Based on D-S Evidence Theory

Tao Liu, Yuanzi Zhou, Junzhong Bao, Xizhao Wang, Pengfei Zhang

Research output: Chapter in Book/Report/Published conference proceedingConference contributionpeer-review

Abstract

In view of the complex environment in which risk analysis of ship accidents is carried out, and the uncertainty and ambiguity of experts’ judgements in risk analysis, this paper proposes a risk analysis method based on intuitionistic fuzzy linguistic set and D-S evidence theory. Intuitionistic fuzzy entropy is applied to determine the weight of each criterion, and then the risk value of m for each attribute is aggregated based on D-S evidence theory. In terms of expert information aggregation, expert weights are firstly obtained based on evidence distance and fuzzy entropy, and then experts’ judgements for attributes are merged via D-S theory to yield the risk ranking of each attribute. Finally, a cruise ship collision scenario is provided as a case to verify the rationality and effectiveness of the proposed method. The result validates that D-S evidence theory is an efficient tool for ship risk analysis.
Original languageEnglish
Title of host publicationNeural Computing for Advanced Applications
Subtitle of host publicationNCAA 2021
EditorsH. Zhang, Z. Yang, Z. Zhang, Z. Wu, T. Hao
PublisherSpringer Singapore
Pages728-741
Number of pages14
ISBN (Electronic)978-981-16-5188-5
ISBN (Print)978-981-16-5187-8
DOIs
Publication statusPublished - 20 Aug 2021
EventNeural Computing for Advanced Applications: Second International Conference - Guangzhou, China
Duration: 27 Aug 202130 Aug 2021

Publication series

NameCommunications in Computer and Information Science
Volume1449

Conference

ConferenceNeural Computing for Advanced Applications
Abbreviated titleNCAA 2021
Country/TerritoryChina
CityGuangzhou
Period27/08/2130/08/21

Fingerprint

Dive into the research topics of 'A Novel Approach to Ship Operational Risk Analysis Based on D-S Evidence Theory'. Together they form a unique fingerprint.

Cite this