可用性观点下的烟叶质量粗糙集评价方法

摘要

By digesting new requirements and influencing factors of tobacco leaf quality in modern cigarette indus-try, a novel comprehensive evaluation index system of tobacco leaf quality under the view of usability is built up. Regarding the complexity and diversity of index data and the particularity of problem through the assessment, we try to extend Pawlak rough set model to a new kind of rough set model based on upper-lower similarity relation, and construct an intelligent rough set model for tobacco leaf quality evaluation under the view of usability in fur-ther.In this way, objective index weights calculated by historical data without prior knowledge and subjective in-dex weights based on expert experiences, are put forward to reach the dynamic understanding of tobacco leaf usa-bility.So, it’ s the first time we’ ve tried to apply and express the concept of usability in a quantitative way.At the end of this paper, a flavorful tobacco leaf selection example is given to prove the feasibility and superiority.%通过对现代卷烟产业中烟叶质量新需求的理解和归纳，构建了可用性观点下的烟叶质量综合评价新指标体系。考虑到实际烟叶质量评价中的数据的复杂性和问题求解的特殊性，将基于等价关系的Pawlak粗糙集模型扩展为基于上、下近似相似关系的扩展粗糙集模型，并设计了相应的可用性观点下烟叶质量粗糙集智能评价模型，在实现了不依赖主观先验信息求取各指标客观权重的同时，进一步引入了专家的主观权重信息，以达到对烟叶“可用性”需求的动态调整。文章首次尝试了基于定量化解决途径来应用和阐释烟叶的“可用性”概念，文末的实证分析验证了本文方法的可行性和一定程度的优越性。