A New Quantitative Approach in the Evaluation of Homogeneity in Deep Mixing-Paper

Paper

Pdf Summary

Deep soil mixing is widely used for ground improvement, but its homogeneity is rarely measured directly. Conventional quality assurance typically relies on strength tests after construction or process indicators such as blade rotation number (BRN). These provide only indirect evidence of mixing quality, and BRN may not apply well to different mixing-tool designs. Specific mixing energy may offer a more comparable production indicator, but requires further study.<br /><br />The paper reviews homogeneity as the uniform distribution of soil and binder at a defined scale, and discusses statistical mixing indices. In practice, measuring binder concentration across enough samples is costly and time-consuming. The authors therefore propose a direct post-production method to quantify untreated soil inclusions in deep-mixed elements.<br /><br />The method involves drilling cores at multiple locations and depths, cutting them lengthwise, and using image processing to distinguish inclusions from the mixed matrix. It measures the proportion and size of inclusions, their variation with depth, and an average inclusion percentage weighted by core length. In a clay-soil trial, more than 80% of identified inclusions had an equivalent diameter below 10 mm, relative to a 2.4 m column diameter and a scrutiny scale of one-tenth that diameter.<br /><br />The approach can help identify poorly mixed zones, assess construction performance, and improve mixing procedures. Its reliability depends on representative, statistically sufficient core sampling and clear image contrast; further validation across different soils and binders is needed.

Keywords

deep soil mixing

ground improvement

homogeneity

quality assurance

blade rotation number

specific mixing energy

mixing indices

binder concentration

image processing

untreated soil inclusions