RT journal article
T1 HADAD: Hexagonal A-Star with Differential Algorithm Designed for weather routing
A1 Jiménez de la Jara, Javier
A1 Precioso Garcelán, Daniel
A1 Bu, Louis
A1 Redondo Neble, María Victoria
A1 Milson, Robert
A1 Ballester-Ripoll, Rafael
A1 Gómez-Ullate Oteiza, David
A2 Ingeniería Informática
A2 Matemáticas
K1 Decarbonization
K1 Hexagonal grid
K1 Optimization
K1 Seasonal study
K1 Variational methods
K1 Weather routing
AB We present HADAD (Hexagonal A-Star with Differential Algorithm Designed for weather routing), a novel optimization algorithm for weather routing. HADAD conducts a global exploration using an A⋆ search on a hexagonal grid with higher-order neighbors, enhancing directional flexibility and overcoming limitations of traditional graph searches that constrain vessel movements. It then refines the solution using a discrete Newton–Jacobi variational method, ensuring convergence to a locally optimal, smooth route in continuous space. To evaluate the effectiveness of HADAD, we developed a benchmark comprising 1,560 instances over a full year, varying in origin–destination pairs, vessel speeds and oceanographic conditions. Our results show that HADAD outperforms pure A⋆ graph search methods by an extra 4% savings with respect to the shortest-distance route, thanks to more flexible smoother trajectories obtained by gradient descent. In our seasonal study we observe that the savings distribution shows large seasonal variations (double savings on average in winter with respect to summer) and contains a significant number of outliers. Savings reach 27% in these cases of extreme weather events. Validation of the algorithm performed with synthetic vector fields has been conducted. In this setting, the algorithm has been adapted to handle fuel consumption optimization for Just-in-Time arrival. By integrating global search and local optimization, HADAD effectively balances computational efficiency with route optimality, offering a practical and adaptable solution for real-world weather routing applications.
PB Elsevier Ltd
SN 0029-8018
YR 2025
FD 2025
LK http://hdl.handle.net/10498/36322
UL http://hdl.handle.net/10498/36322
LA eng
DS Repositorio Institucional de la Universidad de Cádiz
RD 10-may-2026