Multiple-Criteria and Group-Decision Making in the Fleet Selection Problem for a Public Transportation System

The paper presents the methodology of selecting low-floor trams for a medium sized public transportation system in a metropolitan area of Poznan city in Poland. The real world case study of selecting trams in an open bid competition constitutes a background for a research project. The author proposes a universal methodology of solving the decision problem at stake. The proposed methodology has a hybrid character and combines the concepts and rules of two methodologies: MULTIPLE CRITERIA DECISION MAKING/AIDING and GROUP DECISION MAKING. The tram selection problem is formulated as a multiple criteria ranking problem and solved with the application of selected MCDM/A methods (ELECTRE, AHP). The dynamics of individual interactions between members of the groups of Decision Makers (DM-s) and stakeholders as well as between the considered groups (municipal authorities, operator and residents - passengers) are considered. Different approaches to modelling and aggregating the preferences of the DM and stakeholders are presented. The Ex-ante analysis (group interaction prior to computational experiments), focused on building a common preference model for all stakeholders before calculations is compared with the Ex-post analysis featured by a series of computational experiments performed according to different preference models and aggregating its results afterwards (group interaction after computational experiments). A series of computational experiments is carried out and the generated results are demonstrated. They are compared and thoroughly analysed.