Optimizing the Tobacco Leaf Quality Assessment System via Grey Relational and Circle Ill-Condition Index Analysis

Traditional methods for assessing tobacco leaf quality rely heavily on experience and subjective judgment, resulting in inefficiency, and a lack of accuracy and reproducibility. To address these challenges, there is an urgent need for a tobacco leaf quality assessment system capable of assessing various aspects of leaf quality in a systematic and quantitative manner. In this paper, we quantitatively analyze the tobacco leaf quality assessment system for Sichuan tobacco, specifically focusing on the Yunyan 87 variety. In further, we employ Grey Relational Analysis and Circle Ill-Condition Index Analysis for indicator selection and optimization. Through quantitative analysis, key indicators that significantly impact the assessment results and exhibit low information redundancy are retained, and weights are reallocated based on correlation analysis. The optimized system retains four core indicators: Oil Content, Delicacy, Total Nicotine, and Sugar-to-Nicotine Ratio, significantly enhancing the scientific rigor and practical utility of the assessment process. This optimized assessment tool provides a scientific basis for quality control in Sichuan tobacco production, ultimately enhancing production efficiency and product quality.