Machine learning model for delay risk assessment in tall building projects
Risky projects such as tall buildings have suffered an alarming rate of increase in delays and total abandonment. Though numerous delay studies predominate, what is lacking is constructive research to develop tools and techniques to wrestle the inherent problem. Consequently, this paper presents the...
Saved in:
| Main Authors: | , , |
|---|---|
| 格式: | Article |
| 出版: |
Taylor and Francis Ltd.
2020
|
| 主题: | |
| 在线阅读: | http://eprints.utm.my/id/eprint/93117/ http://dx.doi.org/10.1080/15623599.2020.1768326 |
| 标签: |
添加标签
没有标签, 成为第一个标记此记录!
|
| 总结: | Risky projects such as tall buildings have suffered an alarming rate of increase in delays and total abandonment. Though numerous delay studies predominate, what is lacking is constructive research to develop tools and techniques to wrestle the inherent problem. Consequently, this paper presents the development of a machine learning model for delay risk assessment in tall building projects. Initially, 36 delay risk factors were identified from previous literature, and subsequently developed into surveys to determine the likelihood and consequence of the risk factors. Forty-eight useable responses obtained from subject matter experts were used to develop a dataset suitable for machine learning application. K-Nearest Neighbors (KNN), Artificial Neural Networks (ANN), Support Vector Machines (SVM) and Ensemble methods were considered. Feature subset selection revealed that the most relevant independent variables include “slowness in decision making”; “delay in sub-contractors work”; “architects’/structural engineers’ late issuance of instruction”; and “waiting for approval of shop drawings and material samples”. The final results showed that the best model for predicting the risk of delay was based on ANN with a classification accuracy of 93.75%. Ultimately, the model developed in this study could support construction professionals in project risk management of tall buildings. |
|---|