Recently, the big data can be employed as the economical ship operating or evaluation of ship performance conditions. However, such data cannot be easily obtained and analyzed for every ship. In this case, for example, an evaluation of ship performance during operation is usually dependent on ship owner’s experience. The time-dependent ship performance is an essential topic for ship owners because if they realize their current ship performance, they can implement something such as hull or propeller cleaning for their economical operation. This study is focused on the usage of noon report data rather than the big data due to their obtainability. Usually, such data are considered as references because different ship operational condition and environmental condition obscure current ship performance. However, our unique approach, which is used integrally the noon report data such as BHP, propeller revolution and fuel oil consumption, ship sea trial data and propeller performance, can be evaluated ship performance during ship in service. The analyzed output data can be produced as increasing of ship resistance (delta Rw) versus ship performance efficiency, fuel oil consumption (ton per day) or sea margin. Under this output conditions, it can be comparable at same conditions even though the conditions of operations are different. Therefore, this analyzed data has a potential ability to have a look at ship performance conditions during ship in service. The purpose of this paper is to introduce our unique approach using noon data for time-dependent ship performance and then discuss the verification of this approach. As the case study, the noon report data for Japanese domestic bulker was chosen and the ship performance was evaluated in terms of different points of views. It was done comparing the conditions of before and after dry dock to evaluate our approach. In addition, the potential application of this approach will be discussed in this paper.
Partitioning subjects based on high-dimensional fMRI data: comparison of several clustering methods and studying the influence of ICA data reduction in big data