Performance Evaluation of Empirical Equations for Compression Index Using Experimental Data and Statistical Analysis

A quick and economical solution to find the consolidation settlement is to use empirical equations to estimate the soil compression index. These equations may not be appropriate for soils in different regions. In this study, the performance of some empirical equations was determined by estimating the compression index of Kerman clayey soil, Iran. In order to evaluate the compression index, 122 standard odometer tests were performed on undisturbed specimens and the compression indices calculated. Four quantity criteria including error and ranking were selected. The computational measures included determination of the root mean square error (RMSE), ratio of the estimated compression index to the measured value in the laboratory (K), the ranking index (RI) and the ranking distance (RD). Ranking index and ranking distance are related to the accuracy and precision of the estimated results compared to real data. Finally, evaluation of performance of the empirical equations was carried out according to the RD, since it is a self-inclusive statistical parameter that gives accurate and precise estimation compared to other parameters. Results of the analysis indicated that the best estimate of compression index of Kerman soil with empirical equations is based on the initial void ratio compared to other soil properties.