Analysis of pollution prevention performance of vessels in Southeast Asia: Implications towards vessel emission control and reduction

Abstract

As one of the major sources of pollution emissions in Southeast Asia (SEA), substandard vessels with deficiencies in pollution prevention ability may induce extreme results in local societies and cause pollution to marine environments, an issue that needs to be paid close attention to. Through the incorporation of the tree-augmented naïve learning approach and the maximum a posteriori probability estimation approach, a data-driven Bayesian network focusing on vessel performance in the pollution prevention aspect is trained and developed in this research based on the inspection records in SEA from 2017 to 2022 for the first time. A thorough analysis of the proposed model is conducted subsequently to clarify the influence of different factors on vessel performance in pollution prevention in the SEA. This work reveals insightful implications and contributes to the current research in the following ways: First, key risk variables and sub-areas with strong influence on vessel performance of pollution prevention are identified, i.e., place of detention, vessel size, D141-Marpol Annex I, D144-Marpol Annex IV. Second, characteristics and behaviours of substandard vessels in this aspect are identified for better port supervision and inspection. Third, suggestions to control the unaware actions of ship owners and encourage ship owners to endeavour to enhance the pollution prevention ability of vessels are proposed, i.e., formulate targeted instructions on the vessels with identified characteristics of the substandard vessel and against the violation actions of key sub-areas, implement specialised countermeasures for different countries to restrain the occurrence of substandard vessels in their regions.