Resolving the issue of multiple optima in Technology Forecasting using Data Envelopment Analysis

Technology Forecasting with Data Envelopment Analysis (TFDEA) offers an effective means to determine technological capability over time without the burden of fixed a-priori weighting schemes. However, a commonly used extension of TFDEA can create situations where there are multiple feasible solutions. This can introduce challenges in reproducing results for practitioners. This paper resolves this issue by extending TFDEA in a manner consistent with common Data Envelopment methodologies to ensure consistent and reproducible results. The extension is then demonstrated numerically using a well-known data set of fighter jet technology forecasting in which this issue of multiple optima was first observed.