Back to Search
Start Over
Regression Modeling for Prediction of Earned Value Indexes in Public Building Construction Projects: The case of Ethiopia.
- Source :
- Cogent Engineering; 2023, Vol. 10 Issue 1, p1-30, 30p
- Publication Year :
- 2023
-
Abstract
- The construction industry is the sector involved with the repair of buildings and civil engineering structures in an economy. Public building construction projects are constructed repeatedly, and also it becomes difficult to complete projects in the stated period and the estimated cost of the project. Thus, the study was focused on regression modeling for the prediction of earned value indexes in public building construction projects. The regression model was used as a data analysis method to increase the accuracy of the standard earned value analysis method for the evaluation of project performance. The regression model was built on ten sets of building projects database gathering between 2019 and 2022 through the case study method as a data collection instrument from cities, the Ministry of Construction and Housing, consultants, and contractors. The analysis method was made by using a statistical package for social science and Microsoft excels as an analysis tool. It was found that the regression model can predict earned value indexes with the excellent rank of accuracy for the correlation coefficient (R) of the cost performance index model, schedule performance index model, to complete cost performance index model, and to complete schedule performance index model was 88.80%, 95.90%, 96.60%, and 91.40%, respectively, and also for the coefficient of determination (R2) was 78.90%, 91.90%, 93.20%, and 83.40%, respectively. Finally, it can be recommended that the regression analysis method was conducted for its practicality in predicting earned value indexes. [ABSTRACT FROM AUTHOR]
Details
- Language :
- English
- ISSN :
- 23311916
- Volume :
- 10
- Issue :
- 1
- Database :
- Complementary Index
- Journal :
- Cogent Engineering
- Publication Type :
- Academic Journal
- Accession number :
- 174173146
- Full Text :
- https://doi.org/10.1080/23311916.2023.2220497