The effect of the chosen distribution form on reaction probability estimates from drop-weight impact results

The consequences of choosing one distribution form in the analysis of drop-weight impact results were explored, and the importance of reporting both the L50 and the distribution spread-parameter for estimation of reaction probabilities at low-stimuli was demonstrated. The drop-weight impact results of pressed PBX9501 were reported alongside drop-weight results tendered by AFRL. Three distribution forms were fit to the data sets; similar reaction probabilities across all stimulus levels were obtained. It was found that there is little benefit to choosing the normal distribution over the logistic distribution or vice-versa for the sequential test design or analysis of drop-weight impact results.