Theoretical Study on a Statistical Method for the Simple and Reliable Pre-Selection of Wheat Malt Types for Brewing Purposes based on Generally Accepted Quality Characteristics

Germany is the leading malt producing country in Europe. The production capacity for malt was 2.13 million metric tons (t) in 2011, more than half a million t higher than the production in every other country in Europe. According to data provided by the German Malting Association, Germany produced nearly 100,000 t of wheat in 2010 with an increase of 15% in the last two years. According to this, the demand of suitable wheat cultivars for malting and brewing purposes increased. To obtain fast and reliable predications about the suitability of wheat cultivars, a new mathematical method was developed in this work. The method allows a selection based on generally accepted quality characteristics. As selection criteria, the parameters of raw protein, soluble nitrogen, Kolbach index, extract, and viscosity were chosen. The approach is explained by means of one wheat cultivar used in brewing. For this purpose, a complete approach for evaluating measurement uncertainty was used. With the help of simulation calculations using one million measuring data, it could be shown that, on the basis of the arithmetic mean alone, no reasonable prediction could be made about the actual quality of the wheat malt. It is derived step by step how the measurement uncertainty can be determined independently from the sample size by using the standard deviation, the root of the number of samples, and the t-factor. In addition to the interpretation of measurement uncertainty, it has to be considered if the measurand was calculated from other measurands by using a formula and therefore the error propagation must be taken into account. For this reason, the Kolbach index has been split into its components (soluble and total nitrogen), which will likely have an effect on the end result. The estimated error of each individual component was assessed before being combined into an overall uncertainty value. The measurement uncertainty as sample space could be shown as a multidimensional room within two matrices representing the relevant ranges concerning malt quality parameters allowing for a reasonable evaluation of the suitability for brewing.