Applying machine learning and particle swarm optimization for predictive modeling and cost optimization in construction project management
The process of cost optimization in construction projects entails the systematic reduction of expenses and the maximization of value through the efficient management of resources, cost control, and the attainment of project objectives within the limitations of the budget. This study examines the uti...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Published: |
Institute for Ionics
2024
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/105697/ https://www.scopus.com/inward/record.uri?eid=2-s2.0-85165693883&doi=10.1007%2fs42107-023-00843-7&partnerID=40&md5=5cf58a212a77baf3a2055d132957de1b |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The process of cost optimization in construction projects entails the systematic reduction of expenses and the maximization of value through the efficient management of resources, cost control, and the attainment of project objectives within the limitations of the budget. This study examines the utilization of different Machine Learning algorithms, such as Linear Regression, Decision Trees, Support Vector Machines (SVM), Gradient Boosting, Random Forest, K-Nearest Neighbors (KNN), Convolutional Neural Network (CNN) Regression, and Particle Swarm Optimization (PSO), in the domain of predictive modeling and cost optimization in the field of construction project management. The primary aims of this study encompass the improvement of cost estimation precision, the identification of pivotal factors that impact project costs, and the implementation of strategies aimed at reducing costs. Evaluation metrics such as Mean Squared Error, Root Mean Squared Error, Mean Absolute Error, and R-squared are commonly employed in the assessment of Machine Learning models' performance. The Voting regression, which leverages the collective predictive power of multiple models, exhibits superior performance in comparison to individual algorithms. Feature selection methods play a crucial role in identifying the variables that have a significant impact on project costs. By employing these methods, organizations can effectively allocate resources and exercise control over costs. Particle Swarm Optimization (PSO) has demonstrated its efficacy in addressing the issue of construction waste reduction and enhancing the accuracy of cost estimation through the identification of optimal combinations of variables. This study emphasizes the importance of Machine Learning and Particle Swarm Optimization (PSO) in the context of predictive modeling and cost optimization within the field of construction project management. The results of this study can provide valuable insights for professionals in the construction industry, aiding them in making informed decisions, allocating resources effectively, achieving project success, and enhancing profitability. © 2023, The Author(s), under exclusive licence to Springer Nature Switzerland AG. |
|---|