Key points are not available for this paper at this time.
This study investigates the effects of weather condition uncertainties on the propulsion power performance of a cruise ship as a case study along a specific route. Different spatio-temporal ensemble marine weather condition parameters along the given route are adopted to estimate the uncertainties associated with the required propulsion power of the selected ship. First, based on the collected performance data of the considered ship, a feed-forward fully connected artificial neural network (FFNN) is adopted to map the complex relationships between the input data of the weather conditions and the corresponding output propulsion power data. Then, the created FFNN model is fed with all ensemble weather condition members for each grid point of the route to estimate the propulsion power uncertainties. The uncertainties are described using different statistical measures, including standard deviation, box plots, histograms, kernel density estimation, and confidence intervals. Diverse weather data sets are used to quantify the uncertainties by employing the bootstrapping method. The results showed that the weather parameters' uncertainties have considerable effects on the ship's propulsion power, leading to fluctuations in performance and efficiency. These uncertainties can cause variations in fuel consumption, which affects the overall operational costs and environmental impact of the vessel.
Mahmoodi et al. (Tue,) studied this question.