基于历史造价数据的房建工程设计概算编制研究

Research on Budgeting of Housing Engineering Design Budget Based on Historical Cost Data

工程造价咨询业务的数字化转型蓄势待发,利用好长期积累的历史造价数据资源,并挖掘海量存量数据的经济价值是工程造价咨询企业需要面对的问题。文章拟基于大数据分析方法,依据工程造价咨询企业积累的房建工程历史造价数据,以概算指标为例进行挖掘应用,以期能在项目初期快速进行概算编制以助于项目投资控制目标的实现;通过研究确立房建工程概算工程量指标外部环境影响因素体系和内部特征因素体系,进一步基于聚类算法、工程造价计量计价实务及大数据分析理论,构建了基于历史造价数据的房建工程概算指标预测模型;通过以某保障房住宅项目为具体应用分析案例,对类似项目的房建单位工程中混凝土及钢筋混凝土分部工程的分项工程量指标进行叠合并组价,形成初步预测结果,再将预测结果与案例实际值进行对比,根据得到的反馈结果返回修正模型系数以期提高概算指标预测模型输出预测值的准确性,结果证明能够较好地符合预期控制目标。

The digital transformation of social engineering cost consulting business is ready for development.Making good use of long-term accumulated historical cost data resources and mining the economic value of massive stock data are the problems that engineering cost consulting enterprises need to face.Based on the big data analysis method and the historical cost data of housing construction projects accumulated by engineering cost consulting enterprises,this paper intends to take the budget estimate index as an example for mining and application,so as to quickly prepare the budget estimate in the early stage of the project to help realize the project investment control objectives.By studying and establishing the external environment influencing factor system and internal characteristic factor system of the project quantity index of the budget estimate of the building project,and further building the prediction model of the budget estimate index based on the historical cost data based on the clustering algorithm,the engineering cost measurement and valuation practice and the big data analysis theory.By taking an affordable housing project as a specific application analysis case,the project quantity indexes of concrete and reinforced concrete sub-projects in housing construction unit projects of similar projects are supercombined and combined to form preliminary forecast results,and then the predicted results are compared with the actual value of the case.According to the feedback results,the revised model coefficient is returned to improve the accuracy of the output predicted value of the forecast model,and the results show that it can better meet the expected control objectives.