Cost Prediction Model Based on Bayesian Linear Regression
To accurately and quickly predict the cost of highways in the early stages of the project,the Bayesian linear re-gression equation was used to predict the cost.Firstly,the influencing factors of highway cost were identified,and the cost pre-diction index system was established.Secondly,after the unified processing of the collected highway cost data,the cost predic-tion model was established based on the Bayesian linear regression equation,and the prediction effect was compared with the BP neural network prediction model.Finally,Matlab was used for simulation training and prediction.The results show that compared to the BP neural network model,the Bayesian linear regression model has higher prediction accuracy and stability,with a predic-tion error controlled within 5％,a MAPE of 2.29％,and a determination coefficient of 0.925.It can be seen that the Bayesian regression model has high fitting degree and good prediction effect,and the model has good feasibility and applicability,which can be applied to the cost prediction of highway projects.
cost predictionexpresswayBayesian linear regressionBP neural networkMAPE
万方数据
维普
