Developing a CBR System for Marketing Mix Planning and Weighting Method Selection Using Fuzzy AHP

摘要

Case-based reasoning (CBR) solves a problem through retrieving a similar past solution and reusing it in a new situation. In this study, a CBR system was developed for marketing mix planning (MMP) in a steel manufacturing company. In case retrieval, feature weighting is an important component and plays a key role in CBR efficiency. In recent years, several methods have been presented for feature weighting, which have used retrieval accuracy only for performance ranking of weighting methods and ignored other method dimensions. In order to consider other dimensions for weighting method ranking, this study applied a fuzzy AHP method to prioritize weighting methods. In this study, genetic algorithm (GA), rough set theory (RST) and fuzzy inference system (FIS) were used for feature weighting. Moreover, feature weighting based on FIS was introduced in this work for the first time.