Construction and Application of AHP Judgment Matrix with Spatial Factors
The concept behind comprehensive evaluation involves calculating a comprehensive index for all samples based on multidimensional indicators.It aims to depict the relative development levels of each sample through the comparison of these comprehensive indexes.Weight coefficients assigned to each dimension index can denote their respective importance or contribution rate.The primary objective of comprehensive evaluation is to objectively portray the development level and potential of the research subject.Spatial factors play a crucial role in regional development potential.For instance,considering knowledge spillover,the spread of local technology to surrounding areas significantly boosts their development potential.However,when conducting a segmented comprehensive evaluation across different regions,the evaluation of surrounding areas in the aforementioned scenario may be underestimated.Although data-driven methods,including spatial factors in comprehensive evaluation problems,are common,traditional analytic hierarchy process(AHP)overlooks the spatial characteristics of the evaluation object.Integrating spatial factors are introduced into the judgment matrix to establish a spatial AHP method suitable for spatial data structures.Firstly,a spatial feature measurement matrix is constructed from five aspects:spatial correlation,spatial connection,spatial coupling,spatial projection and spatial transfer.Subsequently,the spatial feature measurement matrix is amalgamated into the judgment matrix,presenting the fundamental idea of the spatial AHP weighting method.In contrast to the traditional AHP method,the comprehensive evaluation results of spatial AHP emphasize both local and surrounding regional development.Finally,in the case analysis section,the unique properties of spatial AHP are analyzed distinct from traditional AHP namely,spatial dependence,distance decay,and spatial heterogeneity by manipulating the space weight matrix,space distance threshold,and space division method.The key findings and methodological implications are outlined as follows:spatial AHP not only inherits the objectivity and practicality of traditional AHP,but also establishes a link between the region's development level and that of its surrounding areas.Among the five types of weights Scor-AHP,Stra-AHP,Slin-AHP,Scor-AHP and Stra-AHP exhibit stronger robustness regardless of the linear transformation of the original data and have wider applicability.The comprehensive evaluation results of spatial AHP are influenced by spatial dependence,distance decay,and spatial heterogeneity.The selection of the spatial weight matrix should be problem-oriented,and setting an accurate distance threshold is a prerequisite for effectively using spatial AHP.Additionally,appropriate spatial structure division aids in identifying key drivers of sub-regional development and reflects regional comprehensive index heterogeneity.Spatial factors are introduced into the comprehensive evaluation method for the first time,presents various spatial feature measurement methods,and establishes spatial AHP,thereby advancing the development of spatial comprehensive evaluation methods.
万方数据
维普
