In the production of float glass process, difference or fluctuation in raw material components has a significant impact on the quality of products. Therefore, when dealing with multiple batches of the same raw material during large-scale production mixing, it is necessary to select batches reasonably so that the mean percentage of raw material quality is as close as possible to the designed formula values, thereby minimizing the adverse effects caused by different raw material batches on the entire process. Through the analysis, it was found that with the help of the idea of permutations, it was possible to quickly iterate through all batch combinations of a given stock batch and a specified number of batches, and compared the sum of the mean of the percentage of substances in each batch with the absolute value of the error of the set value, and the batch combination with the smallest absolute value of error can be used as the batch combination used for the final mixture. In this method, the advantages of finding the batch combination closest to the set value are not only absorbed in the selection of batch combinations, but also quickly iterating through all batch combinations to obtain the global optimal solution was also realized.