Applying Group Method of Data Handling (GMDH) Method to Predict Depth to Bedrock

Incorrect depth to bedrock (DTB) estimations can result in a layer moduli backcalculation error of more than 100%. It would be the easiest and least expensive approach to determine the DTB from the falling weight deflectometer (FWD) data itself. This paper demonstrates how to develop a DTB prediction model by use of the group method of data handling (GMDH). Records from 84 locations at 10 sites in Texas were employed to develop the DTB prediction model (70 records for model development and 14 records for model verification). No previous study had collected such a large body of data. The findings show that the R² of the proposed model for the verification data is 0.6635. Given the complexity of geotechnical structures, the model developed in this study provides reasonable DTB prediction. The model and algorithms proposed in this study provide a good foundation for further refinement when additional data becomes available.