An Approach for Risk Prioritization in Construction Projects Using Analytic Network Process and Decision Making Trial and Evaluation Laboratory

Amirhossein Karamoozian, Desheng Wu, C. L. Philip Chen, Cuicui LuoIEEE Access 2019 · OpenReview

Abstract

Construction projects always deal with different types of risks such as financial problems, legal factors, availability of resources and other issues during the project lifecycle. According to the type, size and complexity of the project, the number and intensity of risks could be different and many projects cannot reach the project goals due to exposure to multiple risks. This study aims to propose a hybrid DEMATEL-ANP model for risk prioritization in construction projects and the originality of the work comes from its ability to consider interdependencies between risk factors. The first stage of model applied Delphi method to use experts’ judgments for risk identification. The second stage used Decision-Making Trial and Evaluation Laboratory (DEMATEL) method to define the interdependencies’ relative intensity among the risks. Afterwards, Analytic Network Process (ANP) method is used in the third stage to assess the relative importance of the risk factors to define risk priorities in project. A case study of oil pipeline project is reported to indicate efficiency and performance of the proposed model. The results indicated that the proposed methodology could successfully reveal the important risk factors and define the interdependencies between them in the case study. On the whole, the proposed methodology can be considered as an efficient approach for risk assessment in construction projects.

Introduction

Construction projects are notoriously complex, involving a volatile mix of financial, technical, and political factors. The central challenge for project managers is not just identifying these risks, but prioritizing them effectively. Traditional risk management often utilizes linear models, such as the Analytic Hierarchy Process (AHP), which assume risks are independent. However, in reality, risks are deeply interdependent; a single financial fluctuation can trigger a cascade of execution delays and technical failures. This paper proposes a hybrid framework to map these "butterfly effects" and provide a more realistic assessment of project vulnerability.

Methodology: A Hybrid Decision-Making Framework

To capture the complexity of risk interactions, the authors developed a three-stage hybrid model that integrates qualitative expert consensus with quantitative network analysis.

The proposed three-stage hybrid DEMATEL-ANP framework for risk prioritization.

Stage 1: Risk Identification (Delphi Method)

A panel of 21 industry experts—45% of whom possessed over 20 years of experience—used an iterative survey process to identify 44 potential risk factors for an oil pipeline project. These factors were organized into five primary categories: Technical, Financial, Execution, Management, and Policy.

Stage 2: Interdependency Mapping (DEMATEL)

The Decision-Making Trial and Evaluation Laboratory (DEMATEL) method was used to analyze the cause-effect relationships between risks. By rating how much one risk directly influences another, the model produced a visual diagraph distinguishing between "driver" risks (causes) and "receiver" risks (effects).

Stage 3: Weighting and Prioritization (ANP)

Finally, the Analytic Network Process (ANP) calculated the relative importance of each risk. Unlike traditional hierarchies, ANP allows for inner dependence and feedback within the network. The DEMATEL results were used to prune the network, ensuring the ANP only calculated weights for relationships that actually exist, which significantly reduced the computational burden on the expert panel.

Case Study and Key Insights

The framework was validated through a case study of a real-world oil pipeline construction project. The analysis yielded several critical insights:

  • Financial Dominance: Financial risks were ranked as the highest priority for project success, followed by Technical and Execution risks. The three most significant individual risks were identified as Investment problems (R19), Project closure (R21), and Sanctions (R42).
  • The Maintenance Hub: The model revealed that "Pipeline failure due to poor maintenance" (R7) had the highest interrelationship value. While not necessarily a primary cause, it acts as a convergence point where financial, management, and execution risks meet.
  • Model Stability: Sensitivity analysis showed that the hybrid DEMATEL-ANP model was significantly more stable than using ANP alone. Small changes in expert judgment did not radically flip risk priorities, suggesting the model captures the underlying logic of the system rather than individual bias.

Conclusion

The proposed hybrid methodology successfully identifies critical risk factors and their complex interdependencies. While the model is robust, the authors suggest that future work could incorporate Fuzzy Logic to better handle the linguistic uncertainty of expert judgments. Additionally, applying the model to other sectors, such as high-rise residential or tunneling projects, will further demonstrate its versatility across the construction industry.

Paper Citation: A. Karamoozian, D. Wu, C. L. P. Chen and C. Luo, "An Approach for Risk Prioritization in Construction Projects Using Analytic Network Process and Decision Making Trial and Evaluation Laboratory," in IEEE Access, vol. 7, pp. 159842-159854, 2019, doi: 10.1109/ACCESS.2019.2939067.

Summary updated

Abstract

Construction projects always deal with different types of risks such as financial problems, legal factors, availability of resources and other issues during the project lifecycle. According to the type, size and complexity of the project, the number and intensity of risks could be different and many projects cannot reach the project goals due to exposure to multiple risks. This study aims to propose a hybrid DEMATEL-ANP model for risk prioritization in construction projects and the originality of the work comes from its ability to consider interdependencies between risk factors. The first stage of model applied Delphi method to use experts’ judgments for risk identification. The second stage used Decision-Making Trial and Evaluation Laboratory (DEMATEL) method to define the interdependencies’ relative intensity among the risks. Afterwards, Analytic Network Process (ANP) method is used in the third stage to assess the relative importance of the risk factors to define risk priorities in project. A case study of oil pipeline project is reported to indicate efficiency and performance of the proposed model. The results indicated that the proposed methodology could successfully reveal the important risk factors and define the interdependencies between them in the case study. On the whole, the proposed methodology can be considered as an efficient approach for risk assessment in construction projects.
BibTeX
@inproceedings{karamoozian2019approach,
  author = {Amirhossein Karamoozian and Desheng Wu and C. L. Philip Chen and Cuicui Luo},
  title = {{An Approach for Risk Prioritization in Construction Projects Using Analytic Network Process and Decision Making Trial and Evaluation Laboratory}},
  booktitle = {IEEE Access},
  year = {2019},
  url = {https://openreview.net/forum?id=plNalqKHXK},
}
Figure