Abstract
Many previous studies have developed models for estimating the total cost, whether in the planning stage or the early stage of the project. However, models for estimating the overall risk were proposed in the planning stage only. This paper identifies the factors affecting the overall risk in residential projects at the early stage. The 43 risk factors at the planning stage were identified using a Delphi technique. Experts summarize the 43 risk factors into four factors that can be used to predict the overall risk in the early stage of the project. A multilayer perceptron model with one hidden layer was proposed. The mean absolute error rate for the proposed model was 10%. Risk factors can be used to develop a model to predict the impact of overall risk on project cost at the early stage. The developed model helps stakeholders decide whether the project should continue or be terminated.
Original language | English |
---|---|
Article number | 101586 |
Pages (from-to) | 1-11 |
Number of pages | 11 |
Journal | Ain Shams Engineering Journal |
Volume | 13 |
Issue number | 2 |
Early online date | 25 Sept 2021 |
DOIs | |
Publication status | Published - 1 Mar 2022 |
Keywords
- Artificial Neural Network (ANN)
- Data mining
- Multilayer perceptron
- Overall risk
- Residential projects