A decision support system for ship's energy efficient operation: based on artificial neural network method

Changing crew behaviour towards ongoing operations can yield cost-free energy savings. Identifying the inherent energy-saving opportunities within a real-time environment requires insight into ongoing operations' energy costs. Moreover, achieving energy-efficient performance is a huge challenge that requires a robust decision support system for ship front-end operators.

In this thesis, a non-classical methodology is used to develop a fuel consumption prediction model to support an energy-efficient ship operation decision-support system. In this study, a Decision Support System (DSS) is proposed that consists mainly of two primary components; a ship performance prediction model (PPM) and a ship performance optimization model (POM). In the PPM, Artificial Neural Networks (ANNs) and multi-regression methods were employed. Rather than using traditional noon reports to train the ANN model, intensive datasets were used. For validation and testing, three case studies have been conducted, in order to assess the efficacy of the proposed DSS.