基于FCM和LSSVM的建筑工程造价合理性审核研究

Study on the Rationality Evaluation of Construction Cost Based on FCM Clustering and LSSVM

建筑工程造价进行合理性审核贯穿工程项目管理的全过程,及时发现和控制造价编制的误差,对提高建筑工程投资效益具有重要意义。针对传统造价审核方法的不足,构建建筑工程造价合理性审核模型,确定建筑工程特征指标,采用模糊C均值聚类(FCM)从工程项目大样本数据中识别出与待审项目类似的样本数据作为训练样本;针对造价特征指标与费用的非线性,运用最小二乘支持向量机(LSSVM)构建预测模型进行样本训练;应用预测模型得到待审项目的造价预测区间,通过判断待审项目造价值是否在造价预测区间完成审核。结果表明:FCM聚类能快速有效地识别类似工程样本,审核模型能够提高建筑工程造价审核效率和可靠性。

The rationality evaluation of construction project cost covers the whole process of construction project management.It is of great significance to identify and control the errors of compiled cost in time,which assists in improving the investment benefit of construction projects.This paper aims to address the shortcomings of the foundation and methods for the traditional cost evaluation.First,the construction engineering characteristic index is constructed,and the data similar to the project to be reviewed is identified from the large-sample-size data of the engineering projects by fuzzy C-means clustering,which is used as the training sample.Second,considering the nonlinearity of cost characteristic index and project cost,LSSVM is used to develop a prediction model for sample training.Finally,the prediction model is applied to analyze the cost prediction interval of the project to be reviewed,and the evaluation is completed by judging whether the construction value of the project to be reviewed is within the cost prediction interval.The results show that FCM clustering can identify similar engineering samples quickly and effectively.This evaluation model can improve the efficiency and reliability of construction cost evaluation.