Notes on dispersion

This report describes a method to asses if impact dispersion falls within predefined requirements. It also contains a description of the most common measurements of dispersion and how they relate to each other. The suggested method is designed specifically for small number of firings but can be adjuster for larger numbers. The method can be used to asses if the measured dispersion from a weapon is within pre-defined requirements. The suggested method can be summarized as: Five rounds are fired for each range and a hypothesis test analyses whether the dispersion, with 90 % confidence, is equal to or smaller than the requirements. If this is the case then the next step in the method is taken, and otherwise the result should be regarded as failed. In the next step in the method five more rounds are fired, and based on the result from all ten rounds two new hypothesis tests are preformed to analyse whether the dispersion, with 90 % confidence, is small enough. The first hypothesis test evaluates if the dispersion can be smaller or equal to the requirements. If this is true a second test is preformed, and if it is false the result should be regarded as failed. The second hypothesis test evaluates if the dispersion can be larger or equal to the requirements. If it is likely that this is false then it is probable that the dispersion is smaller than the requirements and the test should be regarded as successful. Otherwise the confidence intervals for the estimated dispersion should be calculated for 50-, 90- and 95-percent confidence and an assessment is done whether further trials are necessary. It should be noted that the suggested method analyses the precision and not the accuracy. I.e. the method does not assess how well the designated target is hit, just if the dispersion is small enough.