In large number of real world dilemmas and applications, especially in industrial areas, efficient processing of the data is a chief condition to solve problems. The constraints relative to the nature of data to be processed, difficult dilemma related to the choice of appropriated processing techniques and allied parameters make complexity reduction a key point on both data and processing levels. In this paper we present an ANN based data driven treelike Multiple Model generator, that we called T-DTS (Treelike Divide To Simplify), able to reduce complexity on both data and processing levels. The efficiency of such approach has been analyzed trough applications dealing with none-linear process identification. Experimental results validating our approach are reported and discussed.