Forecasting the Final Cost of Infrastructure Projects in Syria Using Earned Value Management and Artificial Intelligence

Authors

  • Bassam Hassan
  • Samah Makkieh
  • Nver Titizian

Abstract

Syrian construction projects performance generally suffer from failure in term of cost factor. Monitoring and controlling processes under Earned value management methodology (EVM) are insufficient, especially within reconstruction phase; cause, the complex work environment makes the prediction process based on EVM inaccurate. So this search aimed to improve EVM performance in forecasting final cost of Infrastructure projects using artificial neural networks. Lattakia Ariha highway project was chosen as a case study. The three basic value of EVM were used to obtain parameters which were chosen as inputs to the final cost forecasting network. Then the network was trained on several structures. The structure that corresponding to the smallest error was chosen as the best predictive structure. The training phase showed that the structure consisting of 8 inputs, one hidden layer with 9 nodes represents the optimal final cost forecasting network. Finally, the best structure was tested on 15 samples randomly excluded from corresponding training sets. The test results showed the accuracy of neural networks in prediction.

Published

2020-01-29

How to Cite

1.
Hassan B, Makkieh S, Titizian N. Forecasting the Final Cost of Infrastructure Projects in Syria Using Earned Value Management and Artificial Intelligence. Engineering Sciences Series [Internet]. 2020Jan.29 [cited 2021Mar.3];42(1). Available from: http://journal.tishreen.edu.sy/index.php/engscnc/article/view/9395

Most read articles by the same author(s)

1 2 > >>