Extraction of Bunkering Services from Automatic Identification System Data and Their International Comparisons

Abstract

Despite the increased demand for alternative fuel bunkering and the importance of its base formation, quantitative information or statistics on bunkering are very limited due to data availability. This study aims to develop a quantitative method to extract bunkering operations and analyze and compare the actual bunkering operations, such as bunkering service times and starting times in the port area, by extracting anchored vessels using a clustering method and matching them with bunker barges spatio-temporally. The algorithm also reflects the characteristic behavior of bunker barges, including calling at refineries and bunker barge bases. This study then focuses on bunkering in three port areas and compares their characteristics from various perspectives. The study’s key findings reveal variations in vessel types and service times across three port areas, particularly in Tokyo Bay, where shorter service times are observed for containerships and dry bulk carriers due to high fuel prices. Additionally, it highlights differences in bunkering start times, with Tokyo Bay for daytime operations and Singapore Port for a more balanced distribution throughout the day. Furthermore, bunkering locations differ, with Tokyo Bay and Busan Port having most operations at container terminals, while offshore bunkering is prevalent in Singapore Port.