Using Neural Network Model to Estimate the Optimum Time for Repetitive Construction Projects in Iraq

  • Meervat Altaie College of Engineering / University of Baghdad
  • Abbas M. Borhan College of Engineering / University of Baghdad
Keywords: Repetitive construction projects, Predict Time, Actual Time, Artificial Neural Network, nodes, model

Abstract

The main aim of any successful repetitive construction project management systems is completing the project on accurate time, within the planned budget, and with the required quality limits Planning and scheduling. Artificial Neural Network (ANN) model used to predict the time of repetitive construction project in Iraq depends on historical collecting data adopted from (65) by using thirteen variables. The final results showed strong correlation between actual duration and predict duration there was a strong correlation equal (89.9%) between predicted and observed variables for validation data with testing error (1.51%) and training error (1.32%). It showed from ANNs model there are very good agreement with the actual measurements by found the MAPE and average accuracy percentage equal to (7.812%) and (92.18%) respectively.

Published
2019-01-10
How to Cite
Altaie, M., & Borhan, A. (2019). Using Neural Network Model to Estimate the Optimum Time for Repetitive Construction Projects in Iraq. Association of Arab Universities Journal of Engineering Sciences, 25(5), 100-114. Retrieved from https://jaaru.org/index.php/auisseng/article/view/225