Evolution pathway of process risk of marine traffic with the STAMP model and a genetic algorithm: A simulation of LNG-fueled vessel in-and-out harbor. (1st June 2022)

• Record Type: Journal Article
• Authors: Hu, Shenping; Li, Wenjing; Xi, Yongtao; Li, Wei; Hou, Zhiqiang; Wu, Jianjun; Han, Bin
• Is Part Of: Ocean engineering. Volume 253(2022)
• Publication Date: 2022-06-01
• Subjects: LNG-Fueled vessel -- STAMP -- Process risk -- Genetic algorithm -- Risk evolution -- Marine traffic; Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2022.111133
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: Risk evolution characteristics of an LNG-fueled vessel had been simulated by using the systems-theoretic accident modeling and processes (STAMP) model and a genetic algorithm (GA). First, the STAMP model was established to deconstruct the structure of the risk evolution which was affected by external environmental components. Second, given that the risk performance and maritime traffic develop in the time dimension, and the situation changes in the spatial dimension, a GA was used to provide a timeline for the STAMP model. Third, the coupled STAMP-GA model was applied to the LNG-fueled vessel system. Based on the golden section ratios for levels of risk performance, a Cloud model was introduced to provide input data for the GA model. Last, combined with the data for the LNG-fueled vessel and using a route scenario for a specific anchorage to a designated berth, the risk evolution performance was simulated to explore the risk characteristics of the LNG supply process. The results show that the risk evolution mode of the LNG-fueled vessel follows a bathtub-shaped distribution curve. The novel STAMP-GA model proves to be efficient in analyzing the mutation of risk components and the coupled process in the risk evolution. Highlights: 1.The law of navigational risk evolution was obtained using the STAMP-GA coupling model. 2.The risk evolution mode of LNG-fueled vessels is a disturbed U-shaped curve. The introduction of genetic algorithm obtains refinement of quantitative characterization of risk evolution pathway.