ABSTRACT
Preliminary ship design is an important part of the ship design and a reliable design tool is needed for this stage. The aim of this study was to develop an artificial neural network (ANN) model to predict main particulars of a chemical tanker at preliminary design stage. Deadweight and vessel speed were used as the input layer; and length overall, length between perpendiculars, breadth, draught and freeboard were used as the output layer. The back-propagation learning algorithm with two different variants was used in the network. After training the ANN, the average of mean absolute percentage error value was obtained 4.552%. It is also observed that the correlation coefficients obtained were 0.99921, 0.99775, 0.99537 and 0.9984 for training, validation, test and all data-sets, respectively. The results showed that initial main particulars of chemical tankers are determined within high accuracy levels as compared to the sample ship data.
Disclosure statement
No potential conflict of interest was reported by the authors.