ESTIMATING COMPACT PARAMETERS WITH SIMPLE AND MULTIPLE REGRESSION ANALYZES AND ANALYSIS OF THE MEANING WITH THE F TEST

Abstract

Compaction is the compaction of the moist soil by laying it in layers, rolling it by applying vibration. Compaction is widely used both for the frequency control of road embankments and embankment dams and in other important engineering projects. Therefore, it is important to determine the compaction parameters correctly. The lack of sufficient laboratory equipment at the construction sites and the limited deadline of work increased the importance of correlation equations. If there are sufficient data and results are obtained within tolerable safety limits, it is considered appropriate to use statistics in estimation of compaction parameters. This situation provides a gain both in terms of economy and time. In the study, the compaction parameters were tried to be estimated by subjecting the soil index properties of Adıyaman Balkar, Çelik and Pınaryayla Ponds to simple and multiple regression analyzes. First, the data of the soils belonging to the ponds were used, then, as a result of the sieve analysis, the index properties of fine-grained soils containing only clay and silt were evaluated and compaction parameters were tried to be determined. Simple and multiple regression analyzes were made with the help of the Eview program, where the statistical analysis of the existing data was made, and the statistical significance of the obtained models was examined by performing the F test. It was determined that higher correlations were obtained compared to other sample data, especially by using the data of samples that do not contain sand and gravel material.