Study On PC Residential Construction Cost Estimation Based On ABC-SVM

Taking the road of modernization and realizing the sustainable development of the construction industry are the basic consensus of construction industry.Recently,Precast Concrete(PC)houses meet the concept of sustainable development because of outstanding advantages such as energy saving and environmental protection,reliable quality performance and short construction period.However,problems such as high construction,difficult to control,and investment failure have hindered its development in China.Therefore,it is necessary to accurately estimate the construction cost of the project in the early stage of decision-making,scientifically make investment decisions and target cost control,in order to improve the cost management level of the assembled house and promote the development furtherly.Based on the construction characteristics of PC residential projects,the paper establishes a rapid cost estimation model suitable for PC residential projects to improve the accuracy of cost estimation effectively and provide a basis for refined control of costs.The main research contents are as follows:(1)According to the existing engineering cost data,the paper analyzes the current life cycle cost structure status of PC residential buildings,the significance of researching construction cost estimation is clarified,and a theoretical basis for the practical application of construction cost estimation model are provided.(2)The definition of PC residential construction cost are put forward.According to the analysis of the cost composition and influencing factors of PC residential construction,the estimation indexes of construction cost are summarized from engineering characteristics,structural decoration and construction environment,and the Delphi method is used to modify the index system.Then,the construction cost of estimated indicator system of PC housing is established.A screening method for similar sample engineering is constructed by using grey correlation degree,and the introduction of the entropy weight method better compensates for the lack of gray correlation degree to judge the importance of the index in order to improve the estimation accuracy.(3)Considering the limited sample data available in the PC residential market and the complex cost factors,the support vector machine(SVM)is selected as the basic model for PC residential construction cost by comparing common estimation methods.Aiming at the problem that traditional SVM model parameter setting relies on experience value and subjectivity,Artificial Bee Colony algorithm(ABC)is used to optimize the main parameters of the model.The cost estimate model of PC residential construction is established based on ABC-SVM.The actual engineering case is simulated and analyzed in the paper by using the estimation model built,which verifies the applicability of cost estimation model based on ABC-SVM.The model built in this paper can improve the estimation accuracy of PC residential construction cost effectively,and provide reference for the preliminary investment decision and target cost control of similar projects.