Variability Of The Strength Of Soil-Cement Mixtures-Paper

Paper

Pdf Summary

The paper compares variability in unconfined compressive strength (UCS) of soil-cement produced by three wet soil-mixing techniques: Deep Mixing Method (DMM), Cutter Soil Mixing (CSM), and Trench Remixing Deep Wall (TRD). The analysis covers eight North American projects—three DMM, three CSM, and two TRD—across varied ground conditions and applications. UCS results came from wet-grab and core samples.<br /><br />Strength data were broadly consistent with a lognormal distribution. For wet-grab samples, average coefficients of variation (COV) were 0.36 for DMM, 0.38 for CSM, and 0.22 for TRD. For core samples, the DMM average was 0.36 and the TRD value was 0.28; data were insufficient to compare CSM core results. TRD therefore appeared to produce more consistent strength in this dataset, though the number of projects was limited. DMM showed the widest project-to-project range in variability.<br /><br />The authors caution that differences in soil profiles, binder mixes, equipment, and quality-control and sampling procedures may affect the results. These factors were not standardized or fully evaluated, so the observed differences cannot be attributed solely to mixing technique. Overall, the results fall within previously published ranges and support using established COV values as practical design guidance. The paper calls for more project data to strengthen comparisons and improve design methods that account for ground heterogeneity and construction variability.

Keywords

unconfined compressive strength

soil-cement

deep mixing method

cutter soil mixing

trench remixing deep wall

wet-grab samples

core samples

coefficient of variation

lognormal distribution

ground variability