Abstract
Models used for safely staging divers by computers, software, or tables need correlation and validation against real diving data. The process of correlating diver profiles and DCS outcomes against predictive models and fundamental parameters is presented. Data from computer downloaded profile records is correlated against several well known models, that is, USN, ZHL16, VPM, and RGBM using maximum likelihood statistical techniques. Results are tabulated, LANL DB discussed, risk functions constructed, and implications for diving and divers detailed.
Keywords: Maximum likelihood, Model correlations, Profile data banks, Risk.