We are particularly interested to build and implement a Decision support system with the ability to prioritize certain teams for specific incidents, taking into account the importance of each team that acts in case of emergency. In the work described in [3, 4, 5, 6, 7, 8] we have contributed positively for such objective. Considering the relative importance of each team that acts in emergency context and to sort the list of tasks that should perform all possible orders to be given, a collaborative estimation or forecasting technique that combines independent analysis with the maximum use of feedback was applied to build consensus among experts who interact anonymously (Delphi forecasting) was developed in [5, 8]. The topic of interest was distributed (in a series of rounds) between the participating experts who comment on it and modify the opinion(s) until a certain degree of mutual consensus is reached. In our case, the collection and summary of knowledge of a group of experts from a given area was done through various phases of questionnaires, accompanied by an organized feedback. In [5], we have performed the Delphi method based on [1, 2, 9, 10], that consisted in 3 rounds of completed questionnaires. After the consensus on all issues, could be computed and evaluated an index allowing to associate a number in scale which evaluate the order of teams that must be called to perform a certain task. It were identified which of tasks were the most indicated for each team, and the team that shall be called to respond under a certain order of priority. It can be observed that majority of the tasks are carried out by a SAR brigade team, followed by the Reconnaissance team, the Technical brigade teams, the Logistics brigade and the Medical team respectively. The details about the analysis of results per round are available in [8]. Still in [8] the authors have detailed the tasks associated with priority to each team. In the present work we propose a more detailed computation for the weight of experts experience using hierarchical classification, discriminant analysis and multidimensional scaling. We can classify the experience of each expert evaluating the similarity/distance between the individuals in the group of proposed experts and compare with the number of consensus presented in [8]. In this manuscript we propose an alternative way of weighting the experts experience that contributes to a decision support system capable to prioritize a set of teams for certain disaster incidents envolving maritime issues. The decision support system is still been tested but, with this work, we hope to have given an improvement to its optimization.

Work supported by Portuguese funds - the Center for Computational and Stochastic Mathematics (CEMAT), Portuguese Foundation for Science and Technology (FCT), University of Lisbon, Portugal, project UID/Multi/04621/2019, Center of Naval Research(CINAV).

Decision support system,Expert,Delphi method,questionnaire,catastrophe,hierarchical classification,discriminant analysis,multidimensional scaling