Technology forecasting based on efficiency analysis of systems with interdependent subcomponents using network data envelopment analysis
Technology depends on innovation - but innovative developments are hard to predict. In addition to existing approaches for Technology Forecasting, the use of data envelopment analysis (DEA) provides valuable insights and prediction data. DEA offers a method to evaluate the relative efficiency of analysed entities, so-called Decision Making Units (DMU). Using the efficiency analysis features of DEA in Technology Forecasting enables predictions for future developments based on historic data. While classical DEA analyses the DMUs as closed entities (blackboxes), Network DEA considers the internal structure and the sub-components of the DMUs. This inside view can be leveraged for technology forecasting. Further, If there is no historical data available for innovative new technology, new systems can be seen as composites of existing sub-components with the help of Network DEA. In addition, Network DEA allows the detailed evaluation of the building blocks and their interdependencies. This work develops a model for technology forecasting with Network Data Envelopment Analysis (TFNDEA). Based on existing approaches, a new method is defined to predict technological development with regard to the sub-components using Network Data envelopment Analysis.