Towards a Framework for the Generic Specification of Model-Driven Decision Support Systems: Classification Criteria of Model Relationships

This paper is part of a research project which aims at developing a generic model specification on a conceptual layer in order to reduce the empirically observed high error rate of model-driven Decision Support Systems (DSS), typically based on multi-dimensional, spreadsheet-oriented models. Following the framework of Wand and Weber for conceptual modeling, this paper firstly presents basic elements of an appropriate conceptual modeling grammar, consisting of three criteria to classify all essential problem-structures of typical DSS-domains into eight classes being illustrated by representative managerial examples. Based on this framework, future research aims at developing both a generic, conceptual modeling language, and support tools to design and evaluate model-driven DSS.