A multi-source matching model for static data of ships in Chongqing

The estimation of ship CO
2
emissions based on high-frequency Automatic Identification System (AIS) data, serves as a crucial reference and prerequisite to attain the emission reduction target within the shipping industry in Chongqing. Nevertheless, precise estimation of ship CO
2
emissions depends on detailed vessel information, which may not always be readily available in the vessel parameter database. This challenge is particularly pronounced in inland watersheds, where a substantial amount of static ship data is missing. This study addresses this issue by utilizing a sample database comprising information from 2857 ships passing through or mooring in the Chongqing water area. Regression fitting, graded filling, and regional modularization are combined to construct a multi-source matching model, which could be deployed to supplement the missing static ship data within the study area. The regression fitting technique establishes the relationship between a ship’s dimension, and its main engine power. The graded filling technique is used to discover the relationship between a ship’s length and its main engine power, as well as its length and design speed. The regional modularization technique is utilized to find the relationship between a ship’s location information and its main engine power, as well as its location information and design speed. By employing these multiple methods, the success rate of data matching is improved, consequently enhancing the practicality and reliability of the model. The multi-source matching model for static ship data effectively fills the gaps within the static ship database in Chongqing and resolves the challenge of calculating ship CO
2
emissions within the region.