A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Assessment of Risk Impact on Road Project Using Deep Neural Network
https://orcid.org/http://orcid.org/0000-0003-3306-4548
https://orcid.org/http://orcid.org/0000-0003-1155-4159
Road construction projects are complex endeavors, likewise the associated risks. Thus, identifying and assessing the potential risks associated with road projects is crucial to achieving project success. However, these associated risks are complex and dynamic, and the traditional risk assessment methods cannot accurately assess the risks. Hence, this paper presents a neural network approach by developing a deep neural network (DNN) model for assessing risk impact on project schedule and cost performance of road projects. The proposed DNN model was used to assess and predict the risk impact on 207 road projects in the Republic of Korea (ROK), and the accuracy was compared with a baseline neural network (BNN) and random forest regressor (RFR) models. The results show that the DNN model outperformed the BNN and RFR models, given the lowest mean squared error (MSE) and mean absolute error (MAE) values. The theoretical contribution of this study is related to risk assessment theory and the integration of expert judgment into a DNN model for effective risk assessment of road projects. Also, this study provides the practitioners with a systematic approach to predict accurate project schedule and cost performance considering the significant risk factors while enabling proper allocation of contingency allowance to schedule and cost baselines to guarantee project success.
Assessment of Risk Impact on Road Project Using Deep Neural Network
https://orcid.org/http://orcid.org/0000-0003-3306-4548
https://orcid.org/http://orcid.org/0000-0003-1155-4159
Road construction projects are complex endeavors, likewise the associated risks. Thus, identifying and assessing the potential risks associated with road projects is crucial to achieving project success. However, these associated risks are complex and dynamic, and the traditional risk assessment methods cannot accurately assess the risks. Hence, this paper presents a neural network approach by developing a deep neural network (DNN) model for assessing risk impact on project schedule and cost performance of road projects. The proposed DNN model was used to assess and predict the risk impact on 207 road projects in the Republic of Korea (ROK), and the accuracy was compared with a baseline neural network (BNN) and random forest regressor (RFR) models. The results show that the DNN model outperformed the BNN and RFR models, given the lowest mean squared error (MSE) and mean absolute error (MAE) values. The theoretical contribution of this study is related to risk assessment theory and the integration of expert judgment into a DNN model for effective risk assessment of road projects. Also, this study provides the practitioners with a systematic approach to predict accurate project schedule and cost performance considering the significant risk factors while enabling proper allocation of contingency allowance to schedule and cost baselines to guarantee project success.
Assessment of Risk Impact on Road Project Using Deep Neural Network
https://orcid.org/http://orcid.org/0000-0003-3306-4548
https://orcid.org/http://orcid.org/0000-0003-1155-4159
Road construction projects are complex endeavors, likewise the associated risks. Thus, identifying and assessing the potential risks associated with road projects is crucial to achieving project success. However, these associated risks are complex and dynamic, and the traditional risk assessment methods cannot accurately assess the risks. Hence, this paper presents a neural network approach by developing a deep neural network (DNN) model for assessing risk impact on project schedule and cost performance of road projects. The proposed DNN model was used to assess and predict the risk impact on 207 road projects in the Republic of Korea (ROK), and the accuracy was compared with a baseline neural network (BNN) and random forest regressor (RFR) models. The results show that the DNN model outperformed the BNN and RFR models, given the lowest mean squared error (MSE) and mean absolute error (MAE) values. The theoretical contribution of this study is related to risk assessment theory and the integration of expert judgment into a DNN model for effective risk assessment of road projects. Also, this study provides the practitioners with a systematic approach to predict accurate project schedule and cost performance considering the significant risk factors while enabling proper allocation of contingency allowance to schedule and cost baselines to guarantee project success.
Assessment of Risk Impact on Road Project Using Deep Neural Network
KSCE J Civ Eng
Isah, Muritala Adebayo (author) / Kim, Byung-Soo (author)
KSCE Journal of Civil Engineering ; 26 ; 1014-1023
2022-03-01
10 pages
Article (Journal)
Electronic Resource
English
| British Library Conference Proceedings | 2013
The Western Sydney Orbital Road: Impact Assessment or Project Justification?
| Taylor & Francis Verlag | 2002