This paper presents a new approach to the classification of objects which are described by many features to one of the categories in a predetermined set, of categories. The classification is based on case-based reasoning. Rather than classifying new objects using necessary and sufficient conditions, defined by a set of rules, the knowledge and experience memorised during past classifications are used, Each case comprises the description of the object and the category to which it is assigned. The memorised cases are organised in a case base. In the process of classifying a new object, a case which is the most similar to the new object is retrieved from the case base. The new object is assigned to the same category as the retrieved case. The described approach to classification was a basis for the design of an expert system which helps the decision maker classify a given multicriteria problem to an adequate method for its solving.