论文部分内容阅读
在空间多准则决策中,敏感性分析是衡量决策过程中不确定性的重要方法。层次分析法(AHP)是常用的空间多准则决策方法之一,它的决策结果很大程度上受其成对比较矩阵生成的准则权重的影响,如何量化评估这一影响的大小、如何实现空间多准则决策的可视化,一直是其中的难点。本研究在AHP实现空间多准则决策的过程中,利用OAT(One-at-a-Time)方法依次改变成对比较矩阵元素值来调整权重,从而量化地、可视化地表示决策结果对这些元素值改变的敏感程度。这种新方法能够执行自定义范围内的定量化的动态变化分析,测度不同权重分配情况下,空间决策结果的敏感性;通过设计的敏感性分析量化指标评判敏感性的相对强弱,锁定成对比较矩阵中更需斟酌的元素值;并在GIS平台支持下,实现可视化的空间数据统计及分析。本研究还以嘉兴市北部地区的旅游开发适宜性评价作为范例,来说明该分析方法在空间多准则决策中的可行性。案例分析中,通过对五个准则图层以及相关的成对比较矩阵的分析,利用统计表格、模拟结果图和汇总图表等多种形式,确定了矩阵中敏感性最高的元素,并完成了元素值的最终设定。该方法深入浅出地帮助相关决策者和研究人员更好地理解权重敏感性,从而在空间多准则决策过程中降低不确定性,进行更有效的测度。
Sensitivity analysis is an important measure of uncertainty in the decision-making process in space multi-criteria decision making. Analytic Hierarchy Process (AHP) is one of the most commonly used methods of spatial multi-criteria decision making. Its decision-making results are largely influenced by the weight of criteria generated by its paired comparison matrix. How to quantitatively evaluate the impact of this impact and how to achieve space The visualization of multi-criteria decisions has always been one of the difficulties. In the process of AHP’s multi-criteria decision making in space, the OAT (One-at-a-Time) method is used to change the weights of paired comparison matrix elements one by one so as to quantitatively and visually represent the decision results for these element values Change the degree of sensitivity. This new method can perform quantitative dynamic change analysis within a defined range and measure the sensitivity of spatial decision-making results under different weight distributions. Through the sensitivity analysis of the design, quantitative indicators evaluate the relative strength of sensitivity and lock into Compare the matrix elements that need more consideration, and realize the visualization of spatial data statistics and analysis with the support of GIS platform. This study also takes the evaluation of tourism development suitability in the northern part of Jiaxing City as an example to illustrate the feasibility of this method in spatial multi-criteria decision-making. In the case study, the most sensitive elements of the matrix were identified through the analysis of the five criteria layers and the related paired comparison matrices, using statistical tables, simulation result charts and summary charts, etc., and the elements The final setting of the value. This method can help stakeholders and researchers to understand the weight sensitivity better, so as to reduce the uncertainty and make more effective measurements in the multi-criteria decision making process.