Passive laser barriers are designed to protect humans from risks of laser radiation in the production environment. Manufacturers of such barriers have to guarantee the protective effect. Therefore, experiments are conducted to determine their protection times. For this purpose, some sample based approaches exist but they are not statistically reliable. An approach for the description of the distributions of protection times of passive laser safety barriers is presented to develop a statistical reliable method for the estimation of safe protection times out of few samples. The analysis result of the statistical deviations shows that the measured data can be described by a bimodal approach of two weighted normal distributions.