Uncertainty-informed ship voyage optimization approach for exploiting safety, energy saving and low carbon routes

Voyage optimization has been an effective way to find safe, energy-saving and low carbon routes, but it strongly depends on the accuracy of the weather forecast data that exists the uncertainties which has been influencing the effectiveness and capability of the voyage optimization. In this paper, an uncertainty-informed multi-objective voyage optimization approach is presented. The uncertainty of weather forecast is first explored and modeled by using the historic weather forecast data and the corresponding observations. Then a novel multi-objective voyage optimization model is proposed based on the uncertainty of weather forecast. To solve this uncertainty-informed voyage optimization problem, an uncertainty-based risk penalty function is presented to guide the fitness assessment of the evolutionary algorithm. Finally, the weather forecast uncertainty model was evaluated by using the actual voyage record data, and the superiority and effectiveness of the proposed method were confirmed by comparing it with the conventional state-of-the-art optimization method on the North Atlantic where there were adverse sea conditions. Experimental results show that the optimized route obtained by the proposed method can effectively avoid some underestimated areas where the actual sea states far exceed the forecast data, which is significant for ships to enhance the safety and the energy efficiency.

Automated summary

This paper presents an uncertainty-informed multi-objective voyage optimization approach that explores and models the uncertainty of weather forecast, and proposes a novel optimization model and risk penalty function. The method is proven to be superior and effective through evaluation and comparison with conventional state-of-the-art optimization methods.