A data driven approach for predicting the time-dependent settlement of embankments on soft soils

Engineering predictions of embankment behaviour can be inaccurate, resulting in construction cost and schedule overruns. A key reason for this is the significant influence that personal judgement plays in processing soil test data to determine engineering parameters (eg. Cc, Cr, OCR etc.).
The results of a recent settlement prediction exercise held for a trial embankment constructed in Ballina, Australia, illustrated below, demonstrate the extreme differences in engineering predictions due to subjective soil parameter interpretation.

Developed at the University of Western Australia, CRACA is a new and innovative approach for predicting the time-dependent settlement of embankments on soft soils. Oedometer test data are used directly, avoiding the need to estimate highly subjective soil parameters. CRACA can be applied to embankments with and without prefabricated vertical (wick) drains.

CRACA works in 4 simple steps

Step 1: Upload Oedometer data
Data for each load stage of each oedometer test is uploaded in the form of time (minutes) v settlement (mm) data by pasting the data (using CTRL V) into an excel like table.

The data is automatically processed and a coefficient of consolidation and creep parameter (alpha) is determined for each load stage.

Step 2: Specify the initial and final vertical effective stress profiles
The initial and final vertical effective stress (after full consolidation) are specified next. Again this is done by pasting (using CRTL V) data into an Excel like input table.

To confirm the input, CRACA will display the initial and final vertical effective stress profiles in graphical form.

Step 3: Run the consolidation model
A consolidation model is run next. If Prefabricated Vertical Drains (PVD) (wick drains) are used, then an axisymmetric unit cell can be used to simulate the drainage boundary conditions as illustrated below. If PVDs are not used, a simple one-dimensional consolidation model is used.

The boundary conditions are defined in the input fields pictured below

Coefficient of consolidation values that each oedometer test depth are automatically interpreted based on the initial and final in-situ effective stress values.

The consolidation model can then be run by selecting the green run button on the right. A series of output plots, including contours of the degree of consolidation are provided.

Step 4: Calculate the time-settlement response
A strain profile with depth is obtained using the uploaded oedometer data and the initial and final effective stress profiles. A value primary strain value is automatically evaluated at each oedometer test depth, as well as the top and bottom of the consolidation model. Some extrapolation is necessary if tests are not conducted over the entire depth of the consolidation model and users have the option to modify the automatically interpreted values.

Alpha values are obtained in a similar way

Finally, the settlement is computed. The settlement is divided into primary, creep (secondary) and the total settlement is given (sum of primary and creep). Data can be download to an Excel file.

The figure below show a comparison between the CRACA calculated surface settlement and settlement profile, compare with measured values at the National Field testing Facility in Ballina.