Interpretation and development of top-surface grid in subsurface ground profile using Inverse Distance Weighting (IDW) method for twin tunnel project in Kenny Hill Formation

Abstract

Constrained soil investigation works in tunnel construction projects brings inevitable uncertainties in capturing the exact subsurface profile. This uncertainty leads to misinterpretation of tunnel analysis, which can later cause unreliable prediction of settlements trough in tunnel analysis. The initial stage of interpretation of the subsurface soil profile is generating reliable top-surface grid profile. Increasing the number of boreholes can provide more information for modelling the grid profile. The profile is conventionally developed using borehole information and survey plan, which are interpreted manually and sometimes can be erroneous. This situation proves that the lack of method for data interpretation should be supported with another approach such as Inverse Distance Weighting (IDW). IDW is categorized as a type of deterministic method for multivariate interpolation with a known scattered set of points. The assign values for unknown points will be determined by a weighted average of the values available at the known points. IDW interpolation influenced by weight decreases as the distance increases from the interpolated point. This study focused on the development of top-surface grid for Klang Valley Mass Rapid Transit (KVMRT) project using data from borehole investigation and survey plan along the tunnel alignment. Reduced level of boreholes and information from survey plans that were completed for this project were analyzed. Based on the study results, the Model 2 assigned with exponent 5 for IDW analysis is more preferable for subsurface ground modelling in tunnel analysis for this particular project.

A Machine Learning Technique for Spatial Interpolation of Solar Radiation Observations

“…A higher value of r indicates a larger influence of nearby points on the predicted values. Rojas-Avellaneda ( 2007) suggests an exponent ranging from 1 to 4, whereas a more recent study (Faiz et al, 2019) states an exponent of 5 outperforms an exponent value of 2.…”