基于粗糙集的供应链信息共享评价研究

发布时间：2018-09-07 18:23

【摘要】：随着供应链的不断发展,供应链企业间信息共享变得越来越重要,良好的信息共享能够有效减弱“牛鞭效应”影响,提升供应链整体效能。然而当下缺少合理、客观的信息共享综合评价模型,这使得供应链信息共享程度难以确定。因此需要探讨能够合理、准确的对供应链信息共享程度进行评价的方法。本文对常用的评价方法进行分析发现:粗糙集和模糊集都具备解决不确定信息的能力。模糊集是对知识描述的含糊性进行处理,利用隶属函数来表述信息的不确定,通过对事物的模糊刻画来真实的反应事物的本质;而粗糙集则对对象集中的不可分辨关系进行研究,利用上下近似集合关系来处理信息的不确定性。粗糙集的优势在于不需要与数据无关的额外信息,能通过对数据本身进行处理发掘数据间关系。但粗糙集也有其局限性:单一使用粗糙集并不能有效的处理现实中的信息不确定性问题。因此本文将模糊集和粗糙集相结合并应用于供应链信息共享评价中,主要的研究内容和成果如下:首先,对国内外供应链信息共享有关研究成果进行梳理和分析,给出文章整体思路和研究框架,并介绍和分析了供应链信息共享的内涵。从影响供应链信息共享的因素为切入点,构建一套新的评价指标,为供应链信息共享评价奠定良好的理论基础。其次,为了更加细腻刻画属性模糊性和评价者犹豫性,在直觉模糊集和犹豫模糊集基础上提出了三角模糊数犹豫直觉模糊集,同时提出了三角模糊数犹豫直觉模糊集的运算法则、集成算子、得分函数、精确函数等。本文提出的三角模糊数犹豫直觉模糊集能够实现指标的细腻、多维的刻画,更加符合客观事实和人脑思维。再次,为了更加客观求解评价指标权重,本文在粗糙集正区域数量度量和条件信息熵属性度量基础上提出了混合权重粗糙集模型。并提出了基于模糊刻画和混合权重粗糙集的供应链信息共享综合评价模型,该模型以三角模糊数犹豫直觉模糊集为原始数据,通过混合权重粗糙集模型求解,得到评价指标的客观权重,最后经三角模糊数犹豫直觉模糊集的集成算子求得综合评价值,实现供应链信息共享评价。最后,论文对15个供应链企业进行实证研究。通过问卷调查获取原始的三角模糊数犹豫直觉模糊集型评测数据,借助本文提出的基于模糊刻画和混合权重粗糙集的供应链信息共享综合评价方法,并给出详细实证分析过程,证明了本文所提出的评价方法的合理性和有效性。文末,依据实证结果,提出相应的供应链信息共享程度提升策略。
[Abstract]:With the continuous development of supply chain, information sharing among supply chain enterprises becomes more and more important. Good information sharing can effectively weaken the "bullwhip effect" and improve the overall efficiency of supply chain. However, when there is a lack of reasonable and objective comprehensive evaluation model of information sharing, it is difficult to determine the degree of information sharing in supply chain. Therefore, it is necessary to explore a reasonable and accurate method to evaluate the degree of information sharing in supply chain. Based on the analysis of common evaluation methods, it is found that both rough set and fuzzy set have the ability to solve uncertain information. Fuzzy sets deal with the ambiguity of knowledge description, use membership function to express the uncertainty of information, and reflect the essence of things by fuzzy description of things. Rough set is used to study the indiscernibility relation of object set, and the upper and lower approximation set relation is used to deal with the uncertainty of information. The advantage of rough set is that it does not need additional information independent of data, and it can explore the relationship between data by processing the data itself. But rough set also has its limitation: using rough set alone can not deal with the problem of information uncertainty in reality effectively. Therefore, this paper combines fuzzy sets and rough sets and applies them to the evaluation of supply chain information sharing. The main research contents and results are as follows: firstly, the research results of supply chain information sharing at home and abroad are combed and analyzed. The whole idea and research frame of this paper are given, and the connotation of supply chain information sharing is introduced and analyzed. From the point of view of the factors affecting information sharing in supply chain, a new set of evaluation indexes is constructed, which lays a good theoretical foundation for the evaluation of information sharing in supply chain. Secondly, in order to characterize attribute fuzziness and evaluator's hesitancy, a triangular fuzzy number hesitant intuitionistic fuzzy set is proposed on the basis of intuitionistic fuzzy set and hesitant fuzzy set. At the same time, the algorithm, integration operator, score function and exact function of trigonometric fuzzy number hesitant intuitionistic fuzzy set are proposed. The triangular fuzzy number hesitant intuitionistic fuzzy set proposed in this paper can achieve the delicate and multidimensional characterization of the index, which is more in line with the objective facts and the human brain thinking. Thirdly, in order to solve the evaluation index weight more objectively, this paper proposes a mixed weight rough set model based on the rough set positive region quantity measurement and conditional information entropy attribute measurement. A comprehensive evaluation model of supply chain information sharing based on fuzzy characterization and mixed weight rough set is proposed. The model takes triangular fuzzy number hesitant intuitionistic fuzzy set as raw data and is solved by mixed weight rough set model. The objective weight of the evaluation index is obtained. Finally, the comprehensive evaluation value is obtained by the integration operator of triangular fuzzy number hesitant intuitionistic fuzzy set, and the information sharing evaluation of supply chain is realized. Finally, the paper carries on the empirical research to the 15 supply chain enterprises. The original evaluation data of triangular fuzzy number hesitant intuitionistic fuzzy set are obtained by questionnaire survey. The comprehensive evaluation method of supply chain information sharing based on fuzzy description and mixed weight rough set is proposed in this paper, and the detailed empirical analysis process is given. The rationality and validity of the evaluation method proposed in this paper are proved. Finally, according to the empirical results, the paper puts forward the corresponding strategies to enhance the degree of information sharing in supply chain.
【学位授予单位】：湘潭大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F274

【参考文献】