RETRACTED: A novel method based on clustering and decision-making for construction project portfolio selection

Area of Science:

• Construction Management
• Operations Research

Background:

• Project portfolio selection is crucial for construction organizations but faces challenges due to complexity and resource limitations.
• Effective selection methods are needed to optimize project outcomes and resource allocation.

Purpose of the Study:

• To propose a novel integrated method for project portfolio selection in project-based construction companies.
• To enhance decision-making by combining clustering and multi-criteria evaluation techniques.

Main Methods:

• The K-means algorithm for project clustering.
• The SWARA (Stepwise Weight Assessment Ratio Analysis) method for criteria prioritization.
• The MULTIMOORA (Multi-Objective Optimization by Ratio Analysis) method for project ranking and selection.
• Comparison with the WASPAS (Weighted Aggregated Sum Product Assessment) method for verification.

Main Results:

• 25 construction projects were clustered into 4 groups and ranked using the integrated method.
• The proposed method successfully clustered and ranked project portfolios based on 5 main criteria and 18 sub-criteria.
• Rankings derived from sub-criteria were found to be more preferable due to their detailed consideration of desirability.

Conclusions:

• The integrated K-means, SWARA, and MULTIMOORA method provides an effective approach for project portfolio selection in construction.
• Utilizing detailed sub-criteria in multi-criteria decision-making enhances the accuracy and preference of project portfolio rankings.