Seaport sustainable: use of artificial intelligence to evaluate liquid natural gas utilization in short sea shipping

Abstract

In the present research, a methodology is developed to determine the relationship between the variables that define the use of liquefied natural gas in short sea shipping in Europe, through the use of data-mining techniques. The project takes place in the European space, which includes data from 30 countries, the 28 members of the European Union plus Norway and Iceland. A Bayesian network is constructed with the 35 indicators selected, which are classified into five different categories: international trade and transport, economy and finance, population and social condition, environment and energy, and institutional and political. It is found that capacity of liquefied natural gas regasification terminals under construction and modal distribution of cargo transport by inland waters are the two root nodes of the network. In addition, the variables of transport and international trade and economy and finance become the most important in the decision to implement liquefied natural gas as marine fuel, while those of environment and energy and population and condition are the most dependent on the network.