In response to international fuel oil price fluctuations and the need to reduce greenhouse gas emissions, a meteorological shipping route decision-making system based on an BP neural network is proposed for ship managers to improve their ship operation efficiency while considering economic and environmental factors.

First, seven kinds of operational data, namely ship speed, revolutions per minute (RPM), mean draft, trim, cargo quantity, and influence of wind and waves, are extracted from the log-book and noon report, and the ship's fuel oil consumption is predicted by the BP neural network. A meteorological shipping route decision-making system based on the improved Dijkstra algorithm is then used to obtain the optimal route.

Through the experimental analysis of a 12 335 gross ton multi-purpose vessel on the Yingkou-Incheon route, the goodness of fit between the predicted fuel consumption by BP neural network method and the measured value is 79.97%, the prediction effect is good; and meteorological shipping routes of 15 and 17 kn are obtained by the decision-making system.

The meteorological shipping routes obtained by the decision-making system based on an BP neural network are more accurate and reliable. The results of this study can provide technical support for ship owners and maritime management departments in reducing fuel oil consumption and CO₂ emissions.