Research On Container Port Congestion Evaluation Method Based On AIS Data

Ports are the core hub of the international logistics supply chain,and port congestion can seriously damage the normal operation of the shipping industry and logistics supply chain.When the traffic volume of ships arriving at the port increases,there is a high possibility of congestion at the port,and the port authorities cannot quickly determine the cause of congestion,leading to an intensification of port congestion.Therefore,this study proposes five port congestion indicators to help port authorities quickly identify the causes of congestion,develop and implement targeted operational strategies to alleviate port congestion.In addition,comprehensive evaluation of port congestion level based on congestion indicators can also provide a reasonable basis for shipping companies to arrange shipping schedules and meet the high timeliness needs of cargo owners.This thesis comprehensively considered five factors,including port service capacity,anchorage waiting time,channel congestion within the port,number of anchorage vessels,and port scheduling capacity,to comprehensively evaluate the level of port congestion from multiple perspectives.The specific work content is as follows:(1)In order to explore the different berthing behaviors of ships in port,a ship berthing behavior recognition algorithm is proposed,which preprocesses historical AIS data and uses clustering methods to identify different ship berthing behaviors based on ship position(longitude and latitude),ground speed,and port berth geographic location(longitude and latitude),And defined two types of ship berthing states based on different berthing behaviors,namely the "direct berthing" state and the "anchoring waiting" state.(2)Proposed a framework for evaluating port congestion level: based on the identification algorithm of ship berthing behavior,obtain a dataset of berthing ships.Based on the relevant concepts of congestion in traffic flow theory,define each indicator,calculate and mine the ship berthing dataset,and obtain five port congestion indicators.Based on these five indicators,apply fuzzy comprehensive evaluation based on entropy weight method to evaluate port congestion level,Finally,the congestion level of each port is determined based on the principle of maximum membership.(3)Analyzed the trend,volatility,and correlation of various congestion indicators in the port,explored the specific reasons for port congestion,analyzed port operational capacity,and clarified the independent relationship between each indicator.And analyzed the trend of annual congestion level changes at each port,and explored the rules of annual congestion level changes at each port.The results indicated that there is strong independence between various congestion indicators,and the operational efficiency of Busan Port and Singapore Port is relatively high,which can better cope with port congestion situations.It is worth noting that the Port of Los Angeles has shown a level of "severe congestion" throughout the year,and the degree of belonging to severe congestion is extremely high,indicating a rare occurrence of severe congestion in the Port of Los Angeles.This is consistent with the actual situation of the Port of Los Angeles in 2021,indicating that the congestion levels of each port obtained in this study are highly consistent with the actual congestion situation of each port,further indicating that the framework proposed in this study has practical guidance value.