A rational approach to select the number of field tests required to determine subgrade properties

Abstract

This paper presents a rational approach to answer the questions of how many field tests need to be performed at a test site to determine its subgrade strength properties, and what inferences can be drawn from limited number of field tests. MATLAB simulations were performed to generate various data sets that corresponded to site variability, with the coefficient of variation (CoV) ranging from 10 to 80%. Studies were performed based on coefficient of variation (CoV) and regression analysis. This study provides two important charts for evaluating subgrade properties at test sites: (1) to determine the number of field tests to be performed, after which there would not be any significant change in the inferences drawn from the test results; (2) inferences pertaining to subgrade soil property of a field test site, based on a limited number of tests. In order to demonstrate the applicability of the charts developed through this research, light weight deflectometer (LWD) spot tests that were performed at a test site, were considered. The results were used to draw inferences about the modulus of the subgrade material and the possible variability of the test site, in addition to the maximum number of tests that might be performed.

Keywords:

• Field tests
• material variability
• subgrade property
• Coefficient of Variation (CoV)
• regression analysis