| 摘要: |
| 为解决浅剖数据质量评价指标不一、定量化不足的问题,本文基于层次分析法的原理,结合德尔菲法及模糊评价方法,建立了浅剖质量评价的模糊层次分析技术流程,构建了多层次的浅剖质量评价模型,确定了3个一级指标和14个二级指标,给出了各指标在质量评价中的定量权重,定出了4个质量评价等级。文章应用评价模型对深海及浅海两套浅剖数据集进行了评价,根据评价结果及最大隶属度原则,深海浅剖数据总体质量为中等,浅海浅剖数据总体质量为良好。结果显示,该模型可以有效地将专家主观经验以知识驱动的形式转成定量化评价指标,给出的评价结果客观、可量化,减少了数据质量评价过程中人为主观因素影响和片面性,可为数据进一步应用提供较为准确的应用等级建议。 |
| 关键词: 浅剖数据 模糊层次分析法 质量评价 评价指标 |
| DOI：10.11759/hykx20220728003 |
| 分类号:P717;P229 |
| 基金项目:国家重点研发计划(2022YFC2803600);国家海洋信息中心青年基金项目(202101002) |
|
| Application of fuzzy analytic hierarchy process in the quality evaluation of sub-bottom profile data |
|
KONG Min, WANG Feng-fan, GENG Shan-shan, YU Jia, SHU Yu-ting
|
|
National Marine Data and Information Service, Tianjin 300171, China
|
| Abstract: |
| To address the problems of different quality evaluation indicators and insufficient quantification of sub-bottom profile data, based on the principle of analytic hierarchy process (AHP) and combined with the Delphi and fuzzy evaluation methods, this paper established the fuzzy AHP technology process for sub-bottom profile quality evaluation. The quality evaluation model determined 3 first-level indicators and 14 second-level indicators, gave the quantitative weight of each indicator in the quality evaluation, and set out 4 evaluation levels. In this paper, the evaluation model is used to evaluate two sets of sub-bottom profile data in deep sea and shallow sea contexts. According to the evaluation results and the maximum membership principle, the overall quality of the shallow sea profile data is adjudged as medium, while the overall quality of the deep-sea profile data is adjudged as good. The results showed that the model could convert the subjective experience of experts into quantitative evaluation indicators effectively in a knowledge-driven form. Furthermore, the evaluation results obtained were objective and quantifiable, reducing human subjective factors and one-sidedness in the process of data quality evaluation, and could be used for data quality evaluation, which could provide more accurate application-level recommendations for further application. |
| Key words: sub-bottom profile data fuzzy AHP quality evaluation evaluation indicator |
