Use of Historical Bid Data to Estimate Future Unit Rate for Selected Building Work Items
No Thumbnail Available
Date
2026-05-01
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
This research work used historical bid unit rates of selected items found in most construction projects and have less specification deviation from one another on building projects to create a model that enables estimation of future unit rates. The scope was limited to projects in Ethiopia.Data for a total of 37 projects with 80 building blocks were collected and analyzed. Analysis using multiple linear regression technique was performed between the dependent variables (unit rates) and independent variables which are quantity of the work item; the distance of the project from the capital city of the country; building plot area,; total floor area; number of stories/floors; contract signing date; USD to ETB exchange rate on contract signing date; client type; function of the building and contract type. From the analysis, the regression models predicted the unit rate of C-15 sub structure concrete work with coefficient of determination as high as 0.906. Where, high value of coefficient of determination reveals that the link among outcome and predictor variable is respectable. Whereas, the least accurate model was developed for superstructure C20 concrete unit rate with coefficients of determination of 0.403.The key cost drivers or independent variables that appear in most of the regression models are contract signing date, distance of the project from Addis Ababa, Contract type, number of stories and USD to ETB exchange rate. This thesis has proposed a single formula to estimate unit rate for the selected work items. However, it is highly recommended to include more data for improved accuracy and predictive capability of the models. Therefore, the models developed in this research should serve as a stepping stone for further research on the topic rather than generalizing tool beyond the sample projects.
Description
Keywords
Cost estimation, Historical bid data, Multiple regression analysis