基于灰色关联度的AHP权重矩阵构建方法改进及在农地评价中的应用

合理确定评价指标的权重是准确测度农业土地资源可持续利用能力的关键所在。本文依据灰色关联度对层次分析法(AHP)中判断矩阵的构建方法进行了改进,并获取各项评价指标的权重。在此基础上,分别采用多目标线性加权函数综合评价模型和障碍因子诊断模型,对2001-2012年湖北省枣阳市的农业土地资源可持续利用水平及其制约因素进行评估和辨析。结果显示：基于灰色关联分析法(GRA)改进后的AHP方法在确定指标权重时不仅能有效降低主观赋权的影响,也可避免熵值法等客观赋权法所存在的弊端;地均农业生产总值、人均耕地面积、化肥使用强度、有效灌溉面积、年降水量、人均GDP和农村居民家庭人均纯收入是制约枣阳市农业土地资源可持续利用的最主要因素,但在不同年份各因素的阻碍度表现不一。2001年以来,枣阳市农业土地资源利用的社会经济可持续性水平较高,综合可持续利用水平指数整体上呈增加的态势;但在生态可持续方面,2006年之后不仅明显低于之前年份,而且也显著低于同时期的社会经济可持续性。

A modified Analytic Hierarchy Process method based on Grey Relation Analysis and its application in evaluating sustainability of agricultural land use in Zaoyang City,Hubei Province

Analytic Hierarchy Process (AHP) method has been widely applied to measure sustainability of agricultural land use by determining the weight of each evaluation factor. Delphi method was often used in existing AHP-based weight calculation of the evaluation factors. Delphi method is built on an expert weighting model and therefore difficult to avoid subjective judgment of experts. In this article, a modified AHP method is proposed to calculate the weight of each evaluation factor using the Grey Relation Analysis (GRA) method to solve the problem in using the Delphi method. Using Zaoyang City, a representative area in agricultural land use in northern Hubei Province, China, as the case study area, an evaluation system of sustainable agriculture land use was first established considering a hierarchy of levels including goals, rules, and criteria. The rule level consists of four aspects: agriculture land output sustainability, resource input sustainability, ecological sustainability, and socioeconomic sustainability. Second, experts chose one characteristic factor at the criterion level for each rule element to constitute reference variable sequence and the other factors in the same rule level to constitute comparative variable sequence in GRA. Third, grey relation coefficients were calculated between the characteristic factor and the other factors in the corresponding comparative variable sequence. In constructing a pairwise comparison matrix, experts gave each factor at criterion level a dominant value between 1 and 9 based on the grey relation coefficients. Then calculations were performed to find the satisfactory maximum Eigen value, consistency index CI, consistency ratio CR, and lastly, vectors of weights for factors at the criterion level were obtained, reflecting the relative importance of the various factors to the goal hierarchy in the AHP. Using the weights resulted from this procedure, sustainability of agricultural land use in Zaoyang City was calculated using a Multi-Objective Liner Functions Comprehensive Evaluation (MOLFCV) model, and the obstacles to sustainable agricultural land use in the city were identified. The results show that the sustainability of comprehensive land use of the city increased from 2001 to 2012 while the ecological sustainability of agriculture land resource use was strongly lower than the socioeconomic sustainability since 2006. The reason directly lies in the increasing use of chemical fertilizers and pesticides and the continual decline of annual precipitation in the farmland area. However, it was the lack of awareness and priority of ecological environment protection in the local agricultural land use that resulted in lower ecological sustainability than economic and social sustainability of agricultural land use in Zaoyang City. Moreover, obstacles for sustainable agriculture land use in Zaoyang City included suboptimal total agricultural output per hectare of land, per capita area of cultivated land, amount of fertilizer applied per hectare of cultivated land, effective irrigated area, annual precipitation, per capita GDP, and net income of rural households.