Multiple linear regression analysis and artificial neural networks based decision support system for energy efficiency in shipping

The studies of energy efficiency in shipping have grown in importance in light of recent air pollution developments. Moreover, the Fourth Greenhouse Gas (GHG) Study of the International Maritime Organization (IMO) has also revealed that energy consumption and emissions from maritime transportation still continue to increase considerably. This study aims to reduce air pollution from ships and operational costs in shipping by implementing efficiency measures of voyage management. The methodological approach taken in this study is based on decision support systems (DSS). DSSs have been established with the fuel oil consumption (FOC) prediction methods of Multiple Linear Regression Analysis (MLRA) and Artificial Neural Networks (ANN). The FOC prediction models are created with voyage reports data which includes revolutions per minute (RPM), pitch, mean draft, trim, weather condition, and FOC variables being gathered from voyage reports of 19 container ships. Compatibility values of FOC prediction models are at satisfactory levels (76-90%). The developed models provide a comparison with the performances of MLRA and ANN methods for the prediction of FOC as well as revealing the influences of RPM, trim, ballast, and weather routing optimization techniques on energy efficiency. The results suggest that energy savings may be at 32-37%, 6.5-8%, 7-12%, and 6-8% provided with the optimization of RPM, trim, weather routing, and ballast, respectively.

Automated summary

The studies on energy efficiency in shipping have become increasingly important in response to recent developments in air pollution. This study aims to reduce air pollution and operational costs in shipping through efficiency measures in voyage management. The results suggest that significant energy savings can be achieved by optimizing factors such as RPM, trim, weather routing, and ballast.