Data driven TRL Transition Predictions for Early Technology Development in Defence
This paper proposes the framework of TRL (Technology Readiness Level) transition predictions for early technology development in defense. Though predicting future TRLs is an important planning tool, it has been studied less actively than the other critical issues on TRL, and previous studies mostly have resorted to domain experts. The proposed framework is data-driven and utilises both explanatory and predictive modelling techniques. As a case study, the proposed framework is applied to real technology development data from DTiMS (Defense Technology InforMation Service) which is identified as a key resource. The result of explanatory modelling shows that the two predictor variables, TRL before R&D and project cost, are statistically significant for future TRLs. Also, popular predictive models are fitted and compared with various performance indices using 10-fold cross validation. The two selected predictive models are linear regression and support vector machine models with the lowest prediction errors.
Where otherwise noted, the Articles on this site are licensed under Creative Commons License: CC Attribution-Noncommercial-No Derivative Works 2.5 India