Developing a Detailed Prediction Model for Construction Site Overheads Using Artificial Neural Network in Egypt

Gamal El Din, Amr Aly; Attia, Tarek; Ahmed, Mai Maged

doi:10.21608/erjsh.2023.222575.1188

Developing a Detailed Prediction Model for Construction Site Overheads Using Artificial Neural Network in Egypt

Document Type : Research articles

Authors

¹ Department of Civil Engineering, Faculty of Engineering at Shoubra Benha University, Egypt.

² Housing and Building National Research Center (HBRC), Giza, Egypt

³ Civil Engineering Department, Faculty of Engineering, Fayoum University, Fayoum 63514, Egypt

10.21608/erjsh.2023.222575.1188

Abstract

The estimation of costs is a crucial aspect of construction planning that should be carried out during the initial phases of a project to establish its budget. The term "project overheads" refers to the indirect expenses associated with a project, such as providing general services at the site or plant, including insurance, site accommodation, and other similar costs. The goal of this research is to investigate the variables that affect the precision of site overheads estimation. In this study, the effectiveness of artificial neural networks (ANNs) was examined in addressing the challenge of accurately estimating project overhead costs during the initial stages of building design. The research involved developing a comprehensive prediction model to estimate the percentage of site overhead costs. The primary objective of this paper is to examine the methods used for estimating project overheads in Egypt. This involves identifying factors that affect site overhead costs and developing a detailed prediction model for construction site overhead using artificial neural network in Egypt.For the purpose of achieving the goal of this research, A questionnaire was conducted among construction companies in Egypt. It was found that the most important factors (according to past studies and questionnaire results) were project’s duration, Inflation and Interest rate in the Country and project’s size. ANN model has been developed to estimate the overhead percentage depending on its characteristics. It was found that the use of the ANN model is an effective tool for minimizing the amount of work required to estimate the percentage of on-site overhead costs with greater accuracy.

Keywords