Research On Construction Engineering Cost Prediction Based On Gene Expression Programming

With the rapid development of China’s construction industry,the investment in fixed assets of the whole society has increased year by year.However,due to the long construction period of fixed assets and the large amount involved,if the invest,ent decision is unscientific and the cost control has problems,it will inevitably cause huge losses to the builders and construction parties and even financial institutions.It can be seen how to reasonably determine the project’s investment estimate,make scientific investment decisions,and how to effcctively control the project’s project cost is the key to avoiding losses.For a proposed project,the cost forecast in the early stage of project construction is the basis for analyzing the economic benefits of project investment,the premise for determining the investment limit,and the basis for determining the bidding base in the bidding process.Therefore,exploring a precise and efficient construction project cost forecasting method is of great significance to the investment decision and cost control in the early stage of the project.Firstly,this paper analyzes the status quo of construction project cost prediction at home and abroad,and finds that existing cost prediction methods have theif own advantages and applicability,but they all have ccrtain limitations,thus introducing gene expression programming theory to predict construction project cost;Secondly,the classification sub-item engineering is classified by the primary and secondary factor analysis method,and 15 engineering characteristic factors are obtained by combining the statistical preliminary screening of the factors in the past results.On the basis of this,the importance of the factors is judged by the intuitionistic fuzzy analytic hierarchy process.And use the primary and secondary factor analysis method to classify the factors,and finally select 12 major engineering feature factors as the basis of cost prediction;again,through the analysis of existing cost prediction methods and processes,the overall idea of model construction is obtained,and according to the overall thinking The relevant parameters of the gene expression programming algorithm are set,and the experimental program of the model solving is designed and developed in the Visual Studio programming environment.Finally,the model analysis is carried out through the actual engineering case data,and the characteristic factors are excavated many times in the solving progran.The most between the cost and the unilateral cost Model and predict by analysis and evaluation of the analysis proved that the model has high prediction accuracy,able to pre-investment project estimates,bid prices for determining the tender offer and provide reliable reference.