Optimal Design of Container Ships Geometry Based on Artificial Intelligence Techniques to Reduce Greenhouse Gases Emissions

The ship designers seek to create their geometrical 3D ship models to achieve sustainability targets such as reducing carbon dioxide and being safer. This paper aims to provide a powerful and robust tool for the optimal design of container ship geometry in order to reduce the energy required for propulsion which in turn reduces greenhouse gas emissions. In this paper, the Korea Research Institute of Ships and Ocean (KRISO) Container Ship (KCS) model is considered. The data set of the 3D ship container KCS model’s total resistance is generated by employing the statistical Holtrop-Menan method at different lengths, breadths, draft ranges, and speeds. The collected total resistance data set is trained through Artificial Neural Network (ANN) techniques by using MATLAB-ANN code. The output intelligent data are utilized for creating a General Predictive Smart Model (GPSM). The GPSM is used for creating Graphical User Interface (GUI) software to draw the optimum 3D container model at a specified speed automatically which achieves the lowest total resistance. The developed GUI is observed to provide an automatic determination of the optimal ship geometry.