Developing A Methodology to Improve Tracking of Industrial Projects by Forecasting Earned Duration Index (EDI) Using Artificial Neural Networks (ANN)

Abstract

Construction enterprises produce a huge amount of operational data which are distributed among many databases, but these historical data are not collected and organized in the right way to support project tracking decisions effectively. Earned Duration Management (EDM) is one of the essential techniques in tracking construction project by using different indices to measure the progress and the performance of the project time schedule. The research aims to use artificial neural networks (ANN) to develop a model in order to predict the value of the earned duration index (EDI) at any period of the project using the historical data of company’s projects. The optimum design of the proposed forecasting module has been determined. The statistical parameters used to evaluate the accuracy of the proposed model were Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). The results of this research confirm the ability of artificial neural networks in predicting the value of earned duration index (EDI) at any period of the project life cycle ending up with satisfying results. The study recommends expanding the use of gene expression programming to improve the efficiency of the forecasting models for tracking indices of construction projects, which improve the tracking decisions in order to increase the economic performance of construction enterprises