Modern Research in Decision Making

Abstract In recent decade, humans' ability have rapidly increased in the way of producing and storage of data. By increasing the volume of data stored, the needs for a method by which information and knowledge resources within the data could be analyzed is sensed. Multi-criteria decision making methods and data mining techniques in recent decades helped managers in their decision making process. In this study, a method has been proposed for workshops clustering by combining data mining and MCDM approaches. In the proposed method, first data mining is conducted based on the AHP, K-means and Kohonen neural network approaches, then the performance of designed model is measured by validation indices like SSE and variance between cluster methods in order to determine optimal number of clusters. The food industry has been analyzed as a case study and based on the obtained results, four clusters as the optimal number of clusters have been introduced. The obtained clusters are named based on the variables of population distribution, income, value-added industry and industrial activities in the clusters. Finally, the suggestion is proposed in two sections of practical and research for decision and policy makers in industry and the other researchers in this field.