Route optimisation To make voyage scenario simulations more realistic, it is now also possible to combine Gulliver with a rerouting algorithm. With this algorithm Gulliver can proactively change route and speed based on expected weather and sea state conditions. The rerouting algorithm makes use of a path finding method that is commonly used to find optimal paths in gridded data. The algorithm will search for the optimal route according to a predefined set of limited conditions and cost functions. Cost functions are for instance based on fuel consumption, trip duration or the illness motion rating. Limited conditions could be a ‘simple’ significant wave height but also more complex ones like accelerations or green water on deck.

Captain Decision Mimic Gulliver and route optimisation are combined with a ‘Captain Decision Mimic’ algorithm in SafeTrans. This software package has been developed by MARIN and is used for simulating heavy lift, towed transports and offshore operations. Based on a weather forecast scenario the Captain Decision Mimic algorithm decides when to change course or speed to avoid adverse conditions or to keep on track to the destination. Subsequently, the rerouting algorithm will start searching for the optimum route based on this decision. This could for instance be a route to a safe haven to shelter from adverse weather or a route to the final destination.

To simulate human behaviour, decisions taken by the Captain Decision Mimic algorithm are stochastic. The probabilities of different choices are calculated and used to take a decision. An example of a scenario study with SafeTrans is shown in figure 2. Based on the large number of simulations, accurate probabilities can be obtained for limiting conditions.

Operational profile Nowadays realtime operational data is available for almost every ship in the world. For example, Automated Identification System (AIS) data

shows the actual location, speed and draught with a temporal resolution of minutes. Operational profiles can be created by combining these data with detailed ship design information and high resolution weather and sea state data. Currently, we are investigating machine learning techniques to make our scenario analysis tools more adaptive to these operational profiles. This will improve our scenario simulations. Additionally, it could also be used to compare modelled ship behaviour and performance with actual, observed ship behaviour. This can help shipowners to monitor their ship better and take appropriate decisions.

Figure 2: Example of a SafeTrans scenario study with an overview of all simulations and the resulting wave scatter diagram showing the occurrence in promilles.



Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24