Construction Cost Forecasting Research Based On CBR And Neural Network

With the continuous development of China's social economy,people are no longer confined to the traditional concept of the cost of construction projects into pre-control.People began to increase the degree of concern about the cost of construction,building cost forecast has become a very important task in the construction project.The study of construction cost prediction is not only of great significance to construction engineers,construction projects and the broad masses of the people.Construction cost forecast is an important step before the implementation of construction project,which has great impact on various phases such as planning,designing and management.For planners,designers,construction manager,cost forecast provides a guide in terms of feasibility,design,cost control of the construction project and so on.It helps them make better judgement.Although the traditional methods of prediction have their own advantages,but also have limitations.The linear regression method can not express the complex correlation between the relevant factors in the real project.Case-based reasoning cannot precisely describe the association between construction cost and other features.While artificial neural network is considered powerful and universal,it lacks interpretability for the user.This paper begins with the theoretical basis and the use value of the case-based reasoning and artificial neural network,and analyzes the application and advantages and disadvantages of the respective re-cost forecasting,which lays the foundation for the cost forecast.Based on the case-based reasoning method and the construction cost forecasting software based on the neural network method,this paper studies the combination of neural network method and case-based reasoning method,and then forms some neural network method and case-based reasoning The method has the advantages of improving the stability of the neural network prediction results under various conditions and the explanability of the case-based reasoning method,so as to have better application effect.The experimental results show that the method predicts the method Of the higher accuracy,proved the effectiveness of the method.