Abstract:
In order to better evaluate the physical properties of slim cigarettes, correlation analysis and principal component analysis (PCA) were used to analyze a series of indexes such as cigarette weight, draw resistance, hardness, ventilation rate, tobacco loss from cigarette end, and percentage of loose end cigarette from 200 batches of slim cigarettes. A comprehensive method for evaluating the physical properties of slim cigarettes was established. The results showed that: 1) Three principal components were extracted by PCA (PC1: 32.301%, PC2: 30.623%, PC3: 13.774%), and their cumulative contribution rate of variance was 76.699%. Among them, PC1 reflected the deviation rate of the indexes, PC2 reflected the coefficients of variation of the indexes, and PC3 reflected the tobacco loss from cigarette end and the percentage of loose end cigarette. 2) A cigarette physical property evaluation model was established, and the cigarette physical property scores were classified into four categories: excellent, good, moderate, and poor. 3) Sample 8# had the highest score of 93.02 and belonged to the excellent quality category, while sample 15# had the lowest score of 59.94 and belonged to the poor quality category. Based on the scores of the test samples, the physical properties of slim cigarettes from different batches, machines and shifts could be systematically evaluated.