Italian shipping company Grandi Navi Veloci (GNV) has initiated testing of RINA’s SERTICA Performance, a machine learning and predictive model system, on the GNV Polaris vessel to optimise energy consumption.
The system was put into action during the ship’s maiden voyage from China to Italy.
Go deeper with GlobalData
The SERTICA Performance system is designed to monitor operational data on ships, facilitating “efficient” energy management and performance optimisation.
The system has enabled GNV to discover two optimal operating scenarios that achieve the lowest specific fuel consumption, stated the company.
Additionally, a predictive model has been developed, acting as a benchmark and simulator for future operations.
This system operates as a real-time data collector, utilising onboard sensors to record key parameters such as fuel consumption and the power of diesel generators and engines.
While its primary function is energy monitoring, the system also calculates the ship’s actual efficiency, providing a complete picture of operational performance to both the crew and onshore management.
During the GNV Polaris’ journey, various operational scenarios were tested at different speeds and configurations to determine the most fuel-efficient solutions. These included alternating the use of diesel generators and shaft generators.
The system’s accuracy was further validated on the Genoa-Palermo route, where the sea trial results aligned with the forecasts. This confirmed the optimal configuration for reducing fuel consumption, according to the company.
GNV energy efficiency director Ivana Melillo said: “GNV is making significant strides in sustainable shipping with their latest initiatives. One of the most notable developments is the introduction of the GNV Polaris, the first of four new ships designed to enhance sustainability in maritime transport.
“The GNV Polaris boasts high environmental standards and can achieve over 30% fuel savings, resulting in a significant reduction in CO₂ emissions compared to the vessels currently in the fleet.”
The project also entails the creation of predictive models using machine learning techniques.
These performance models can be used to assess performance degradation over time or simulate different operational scenarios.
With more than 800 ships already equipped with SERTICA Performance, its adoption continues to grow, including on vessels currently under construction.
