A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Time Prediction for Highway Pavement Projects Using Regression Analysis
In the planning stages of highway projects sufficient information may not be available to apply a full scale critical path method (CPM) analysis to predict the duration and the use of approximate methods would be required. Highway agencies usually record representative information about the projects, e.g. planned and actual duration, quantities of work, project cost, and project miles. By applying a set of statistical methods to these historical records, a prediction of the duration of a project could be obtained. This paper describes the development of time prediction models through the application of general multiple regression analysis, ridge regression analysis, and nonlinear partial least-square regression analysis to project data from Washington State Department of Transportation. The models were developed using representative variables including the hot mix asphalt (HMA) quantity, grading quantities, and surfacing quantity, the number of project miles, and the contract value. Cluster analysis was used to further improve the prediction ability of the models. Through an example, the developed models were validated against an actual project data. The prediction of project duration is an important objective of highway agencies and the developed models would be helpful tools to achieve this objective.
Time Prediction for Highway Pavement Projects Using Regression Analysis
In the planning stages of highway projects sufficient information may not be available to apply a full scale critical path method (CPM) analysis to predict the duration and the use of approximate methods would be required. Highway agencies usually record representative information about the projects, e.g. planned and actual duration, quantities of work, project cost, and project miles. By applying a set of statistical methods to these historical records, a prediction of the duration of a project could be obtained. This paper describes the development of time prediction models through the application of general multiple regression analysis, ridge regression analysis, and nonlinear partial least-square regression analysis to project data from Washington State Department of Transportation. The models were developed using representative variables including the hot mix asphalt (HMA) quantity, grading quantities, and surfacing quantity, the number of project miles, and the contract value. Cluster analysis was used to further improve the prediction ability of the models. Through an example, the developed models were validated against an actual project data. The prediction of project duration is an important objective of highway agencies and the developed models would be helpful tools to achieve this objective.
Time Prediction for Highway Pavement Projects Using Regression Analysis
Aziz, Ahmed M. Abdel (author)
Construction Research Congress 2009 ; 2009 ; Seattle, Washington, United States
Building a Sustainable Future ; 896-905
2009-04-01
Conference paper
Electronic Resource
English
Time Prediction for Highway Pavement Projects Using Regression Analysis
| British Library Conference Proceedings | 2009
Cost Prediction Of Highway Projects Using Regression
| British Library Conference Proceedings | 2007
Spatial Clustering for Determining Economical Highway Pavement Let Projects
| British Library Conference Proceedings | 2006
Road user and mitigation costs in highway pavement projects
| TIBKAT | 1999