Pilot 4: Weather Routing and Fleet Intelligence Service in Shipping

This pilot aims at enhancing fleet operation optimizations by pursuing several goals on both vessel and fleet scale: a higher accuracy in vessel performance simulation, a proactive adjustment of the voyage plan to respond to variations in the weather forecast, and increased awareness and prediction of ship operations and waiting time. Exploiting models and services developed in VesselAI project NAPA could evaluate the benefits of utilizing artificial intelligence techniques to improve the offering to the maritime industry by supporting and complementing classical methodologies in different aspects, including accuracy, computational time and user experience, and scalability.

Watch the video

Use Cases

During the first evaluation period of the project, the focus of the pilot was on the following use cases

Voyage Optimisation

Shipping companies and operators are supported in all phases of voyage planning, both onboard and ashore, with a simple and intuitive user interface to handle workflow and information exchange and to enable actions at each step. This starts with voyage optimization with improved accuracy in the estimation of costs and voyage duration

Real-time Monitoring of Fleet Operations

Real-time monitoring of fleet operations send notifications of potential deviations due to dynamically evolving conditions, and perform comparisons with alternative proposals en route, enabling a holistic approach to optimize the operations of the whole fleet by improving scheduling thanks to increased awareness and prediction of time spent at berth and waiting at anchorage.

Pilot-participating Partners

NAPA
MarineTraffic
UPRC
VTT

Pilot Technological Highlights

The technological highlights of this pilot were the following

Great attention is given to the scalability of the solutions globally. Consequently, the models require to consume large amounts of data of different types, including:
  • World fleet database for ship characteristics
  • AIS (Automatic Identification System) for tracking vessels' operations
  • Weather forecasts and bathymetry
  • Daily reports and high-frequency sensors' measurements to monitor performances and improve predictions
  • Information about safe navigation passages in restricted waters
Data
Proactive re-optimization along the voyage is realized by two components. A backend task runs the weather routing calculations either at a regular frequency or when needed according to the developed AI model, which will be queried through a dedicated API from the VesselAI platform. A user interface element that enables the user to monitor the operations, possible deviations, and more favorable alternative voyages in real time. To ensure an up-to-date situation awareness about port status, port and terminal layouts are reconstructed quarterly worldwide. Service "Port Mooring Area Clustering" can be used to generate density-based clusters. Considering the massive data flow, both the AIS treatment and polygon cluster creation are implemented in the NAPA cloud.
Components
Regular optimization and adjustment to the voyage plan provide an average of 2% additional savings in fuel consumption and thus emissions savings on longer voyages experiencing harsh weather– on top of the ca. 10% savings from weather routing. AI models developed cut the required computational resources by more than 95%, triggering re-optimization only when required, enabling scalability on the world fleet scale. Moreover, a promising above 10% improved accuracy in the estimation of performance has been tested, with a greater effect in conditions for which classical models are known to be less reliable, e.g., navigation in shallow water.
Highlights

Pilot Results

Advanced clustering techniques of AIS data in and near ports allowed us to reconstruct ports and terminals layouts more quickly and robustly, enabling a reliable estimation and prediction of time spent at berth and waiting at anchorage. VesselAI Pilot 4's focus on weather routing and fleet intelligence service brings promising solutions to the shipping industry. By adopting these operational methods, companies can significantly reduce fuel consumption, cut emissions, and embrace a more sustainable future.