International Journal of Hydrogen Energy, Vol.41, No.47, 22192-22205, 2016
Hydrogen pipelines: Enhancing information visualization and statistical tests for global sensitivity analysis when evaluating multidimensional risks to support decision-making
Using pipelines to transport fuels is considered safe and financially viable. Multicriteria decision models are used as they lead to managing risk better and to more effective decision-making when assessing risk to hydrogen pipelines. Nevertheless, there is recurring concern about the possibility of accidents occurring due to failures in the storage structure and/or when transporting hydrogen. Given the uncertainties inherent when using these models, Monte Carlo simulations are conducted and the results are analysed based on Kendalls ti correlation coefficient. This coefficient was fundamental for understanding the behaviour of different rankings obtained by simulations, since information was generated on ranges of uncertainty and on recommendations to decision-makers. The article suggests steps on how best to conduct a sensitivity analysis and to ensure that results are robust, based on a model that uses graphs within Information visualization that support the decision-maker when ranking pipeline sections and to mitigate risks. (C) 2016 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Keywords:Multidimensional risk analysis;Hydrogen pipeline;Multicriteria decision model;Monte Carlo simulation;Kendall's correlation coefficient;Information visualization