Prediction of construction material prices using ARIMA and multiple regression models

Abstract

Abstract Construction Material Prices (CMP) variations have become a major issue in properly budgeting construction projects. Inability to accurately forecast CMP volatility can also lead to price overestimation or underestimation. Enhancing the accuracy of predictions of CMP can also enhance the accuracy of predictions of total construction costs. The purpose of this study is to present a model for predicting construction material prices that assist decision-makers to make better decisions over the life cycle of a project. The price records for CMP namely; steel, cement, brick, ceramic, and gravel, and the indicators affecting them in Egypt were used for the prediction procedures. The practical methods for using the Box-Jenkins approach Autoregressive Integrated Moving Average (ARIMA) time series and multiple regression models for forecasting building material prices are outlined in this research. Out-of-sample predictions are used to evaluate the provided model's performance in predicting future prices. The models are compared according to the Mean Absolute Percentage Errors (MAPE). The generated models show good results in predicting month-to-month variations in material prices, with MAPE ranging from 1.4 to 2.8 percent for the selected models. This research can assist both owners and contractors in improving their budgeting processes, and preparing more accurate cost estimates.