论文部分内容阅读
高速公路质量检测过程中发现施工单位的内业资料尤其是质量保证资料受重视程度不够,施工自检数据问题较多,可信度不高。本文总结了质量保证资料存在的主要问题,分析了产生原因。将经验判断和数学分析相结合,从试验检测数据的逻辑联系和正态分布等特点出发,综合异常数据识别、统计检验等数据验证方法,提出试验检测数据验证一般流程。选用肖维勒准则法(Chauvenet)分析并剔除异常数据,基于SPSS数据分析软件下的K-S检验、F检验、t检验和Mann-Whitney U检验4个工具依次进行检验,确保施工单位自检数据可靠并能真实反映工程实际质量状况,防止虚假数据蔓延对建设项目工程质量造成损害。
During the highway quality inspection, it is found that the construction unit’s internal information, especially the quality assurance data, is not given enough attention. There are many problems in the construction self-test data and the credibility is not high. This article summarizes the main problems of quality assurance data and analyzes the causes. Based on the combination of experience and mathematical analysis, based on the logical connection and normal distribution of the test data, this paper synthesizes the data verification methods such as identification of anomaly data and statistical test, and proposes the general flow of data testing. The Chauvenet method was used to analyze and eliminate the abnormal data. Based on the SPSS data analysis software KS test, F test, t test and Mann-Whitney U test four tools were tested in turn to ensure that the construction unit self-test data is reliable And can truly reflect the actual quality of the project to prevent the spread of false data on the construction project quality damage.