Research On The Index Weight Of Urban Logistics Node Attraction Degree

Construction of the logistics parks is under way like a raging fire in China.By 2020,a national network of logistics parks with reasonable layout,moderate scale,complete functions and high efficiency will be basically in place.However,there is no quantitative comparison standard for which logistics park is more likely to attract other nodes to develop logistics activities in the whole logistics network.Therefore,this thesis introduces the concept of logistics node degree,which is quantified by mathematical methods and expressed with specific numerical values.This thesis aims to calculate the urban logistics node attraction degree by mathematical methods.The thesis introduces the definition of node attraction and node attraction degree,the relationship between them,the model of logistics node attraction,and determines several principles in the quantitative comparison of attraction degree of logistics nodes in different cities.Firstly,32 indexes related to the logistics node attraction degree are listed,then 15 influencing indexes of logistics node attraction degree are extracted by using factor analysis and principal component analysis,and an index system of logistics node attraction degree is established.The improved normalized transformation formula is used to process the data,and five index weighting methods are used to determine the index weights.Objective weighting method includes entropy valuation method,gray correlation degree method and variation coefficient method,subjective weighting method is analytic hierarchy process,and combination weighting method is arithmetic average weighting method.Then the attraction degree of 9 first-level logistics nodes,46 second-level logistics nodes and 232 third-level logistics nodes are calculated.Combined with the development of logistics nodes in Shenzhen,this thesis analyzes the similarities and differences of the five weighting methods based on the calculation results of logistics node attraction degree,and compares them with the actual situation to judge the advantages and disadvantages of the five weighting methods.For first-level nodes,the attraction degree results of entropy value method and analytic hierarchy process are more practical.For the second and third level nodes,the attraction degree results of the five weighting methods are all in good agreement with the reality.At last,this thesis gives the summary,the innovation and the improvement needed in the future.