In some types of networks, known as influence diagrams (ID), the anti-PD-1 monoclonal antibody decisions are represented by distinctive decision nodes (DNs) that often

are guided by the reaction of utility nodes (UNs) to the network. These two types of nodes (DNs, UNs) are used to automatically help determine the decision to make, which gains the highest expected utility (EU), considering the given circumstances. For the purpose of this study, an influence diagram is used as a way to transmit our knowledge about an analyzed system, its components and their behavior. The use of an ID to develop the cost model allows us to easily determine the oil-combating actions that minimize the total cost of the clean-up operation. The presented model has been developed with the use of Hugin Researcher 7.8 modeling environment. In order to gather data for the model, both literature sources and expert opinions are utilized. Additionally,

some of the conditional probabilities needed for the cost model have already been estimated in previous studies regarding the environmental impact of an oil accident in the Gulf of Finland, see for example Lehikoinen et al., 2013, Partila, 2010, Juntunen, 2005 and Juntunen et al., 2005. Usually, ISRIB in vitro when expert solicitation is used as a way of collecting data for BBNs, one should first decide if the expert will be asked to provide both the model structure and the probability distributions, or if expert knowledge is only to be used for the latter. In the case presented in this paper, the structure of the model is based on the literature review of existing cost models and factors affecting the cost of the clean-up operations. Therefore, the expert solicitation is needed to provide the missing probability distributions, which are not mentioned in existing studies, and to verify the data from previous studies when feasible. In the event of some necessary data not being available, neither in literature

and previous studies, nor by expert solicitation, some generalizations and simplifications have to be made. The expert solicitation for this study is completed primarily during a one-day workshop, where fifteen Molecular motor professionals in the field of environmental issues are gathered. As the information regarding the needed costs and oil-combating ships efficiency is highly dependent on a wide range of situational circumstances, the group discussion approach is preferred to the individual interview scenario. This allows the interchanging of opinions between the interview subjects, and, in some cases, leads to a more accurate result than in the case of personal estimations. A cost model of oil spill clean-up operations, which is developed here, is depicted in Fig. 2. It consists of four types of variables, connected in a logical way. The variable types are as follows: • utility variables; The utility variables represent two groups of costs that are encompassed by the presented model.