Evaluation of spatial variability of core strength from deep mixing columns

Abstract

Quality evaluation of the jet grouting method is normally performed based on the unconfined compression strength of specimens taken from the constructed ground improvement by core boring. The core strength data is the sample data representing the strength of the ground improvement. The core strength varies spatially even under the same soil and construction conditions. Therefore, statistical variables such as mean values and standard deviations of the core strengths are used for the evaluation of the quality of the improved ground. This paper reports the results of an analysis of core strength data obtained from cement-treated soil columns by the jet grouting method. The mean value, standard deviation, and autocorrelation distance of the unconfined compressive strength were evaluated based on the core strength obtained at construction sites. Autocorrelation distance is a variable that allows for the quantitative evaluation of the spatial correlation of strength. The autocorrelation distance was calculated using the maximum likelihood method, assuming that the core strength follows a normal distribution. The Kolmogorov-Smirnov (K-S) test was performed to confirm that the probability distribution of the core strength follows a normal distribution. Based on the K-S test result, the goodness-of-fit for the normal and lognormal distribution against the probability distribution of the core strength was examined. Finally, the autocorrelation distance of the core strength data obtained from soil-cement columns by mechanical mixing method was compared with that evaluated in this study.