The utility variables also dictate the states of the decision variables in such a way that the total costs are minimized. This means that the model can determine the oil-combating strategy, which minimizes the clean-up costs. However, the remaining effects of the oil spill on the environment and society are not considered in this study, and thus, the proposed strategy shall by no means be considered optimal. The decision nodes in the model consist of booms and oil-combating vessels. These nodes only exist in Boolean states of being sent or not sent to the location of the accident. These decision nodes directly affect the offshore clean-up costs and, indirectly,

the onshore clean-up costs. The decision find protocol node Booms refers to the use of offshore booms, with the aim of keeping the oil close to the oil combating vessels for as long as possible thereby decreasing its spreading rate. The use of onshore booms is not anticipated in this model. When it comes to oil combating fleet, the decision nodes account for the three largest and the most effective oil-combating vessels in the Finnish Navy: Louhi, Halli and Hylje. There are also two combined nodes encapsulating smaller oil-combating vessels managed by the state-owned company Meritaito Ltd., and ships belonging to the Finnish Border Guard. This division is justified by the fact that the ships owned by the Finnish Border

Guard and Meritaito Ltd., are rather small and mostly used in

the early stages of the clean-up process, before the larger combating vessels reach the spill location. These ships are grouped into Mitomycin C mw two decision nodes in the model. The node Finnish Border Guard refers to three vessels: Uisko, Tursas and Merikarhu, and the Progesterone node Meritaito Ltd. refers to four vessels: Oili I, Oili II, Oili III and Seili. Due to the size limitations of the model, it is not feasible to include all vessels separately. We also assume that all the vessels belonging to a decision node are sent to combat the spill if the node is selected. The independent variables of the cost model are: Spill size, Season, Oil type and Time for spill to reach shore. The last is more realistic and more useful from the modeling perspective when expressing the distance from the location of the oil spill to the nearest shore. The independent variables allow users to define them, however the model gives an opportunity to select the closest interval from the pre-set states for the node. In the event that these values are not known, the initial variables have their own probability tables and values, obtained in the course of simulations for the environmental and traffic conditions prevailing in the Gulf of Finland. The length of polluted coast is not considered in the model, instead we determine the clean-up costs based on the amount of pollution that reach the shore. Spill size is an independent variable with 10 states, as presented in Table 1.