ABSTRACT
It has been possible to validate to a certain extent the quantitative aspects of human error analysis. To do this, a comparison was made between the predictions made using the method in the 1990s and the actual accident frequencies since then. There are just a few plants where this is possible. It proved impossible to validate the original accident frequency calculations, because the data documented in this book were not available at that time. It was possible to substitute the new data back into the old calculations and to derive new frequencies. The results of the comparisons are given in Ref. [2].
