Comparing Two Optimization Approaches for Ship Weather Routing Laura Walther, Srikanth Shetty, Anisa Rizvanolli and Carlos Jahn Abstract Weather routing in maritime shipping is related to a shipping company’s objective to achieving maximum eﬃciency, economy and cost competitiveness by optimizing each voyage of a ship. A voyage can be optimized regarding cost, time, safety or a combination of these factors, while considering forecasted meteorologi- cal and oceanographic information as well as constraints given by geographic condi- tions, ship characteristics, emission regulations, safety requirements or time restric- tions. A wide variety of mathematical models of the ship weather routing problem as well as diﬀerent approaches to solve it can be found in the literature and are applied by numerous software systems. This paper presents two approaches to solve the ship weather routing problem, a graph algorithm and an evolutionary approach. Both approaches aim to minimize fuel costs, allowing for route and speed optimization. They are compared based on numerical examples with real-world data. 1 Ship Weather Routing Problem Voyage planning and optimization represents a widespread measure to improve cost and energy eﬃciency of maritime shipping. Ship weather routing generally aims to ﬁnd an optimal route and speed proﬁle for a ship’s voyage based on the analysis of metocean weather forecasts. Meteorological institutes commonly use the mathe- matically concise data format GRIB (General Regularly-distributed Information in Binary form) to store weather data numerically predicted for each node of a grid. The ship weather routing problem is mathematically modeled in various ways [16]. L. Walther ( ✉ ) ⋅ S. Shetty ⋅ A. Rizvanolli ⋅ C. Jahn Fraunhofer CML, Hamburg, Germany e-mail: Laura.Walther@cml.fraunhofer.de S. Shetty e-mail: Srikanth.Shetty@cml.fraunhofer.de A. Rizvanolli e-mail: Anisa.Rizvanolli@cml.fraunhofer.de C. Jahn e-mail: Carlos.Jahn@cml.fraunhofer.de © Springer International Publishing AG 2018 A. Fink et al. (eds.), Operations Research Proceedings 2016, Operations Research Proceedings, DOI 10.1007/978-3-319-55702-1_45 337