Spatial Statistical and Multivariate Regression Approach to Earthwork Shrinkage-Factor Calculation

• Main Author: Shamo, Benedict
• Format: Others
• Published: North Dakota State University 2017
• Online Access: https://hdl.handle.net/10365/26924

In this research, a linear shrinkage factor model, which is the result of spatial and statistical modeling, was developed for the state of North Dakota. The input variables for the developed shrinkage factor models were derived from spatially modeled soil data which makes the function responsive to soil variability across the state. The current approach for selecting the shrinkage correction factor in earthwork contracts across the state of North Dakota is through a trial-and-error system. This deterministic system employs the judgment of experienced engineers in selecting a shrinkage factor value for earthwork contracts. The current approach assumes shrinkage factor uniformity and does not provide a measure of the estimate's reliability. Due to the heterogeneous nature of soil properties across the state, the trial-and-error approach for selecting the shrinkage factor greatly impacts earthwork volumes, which could lead to contract variations and increase the cost of contract administration. === North Dakota Department of Transportation (NDDOT)