Artificial Intelligence for Prediction of Physical and Mechanical Properties of Stabilized Soil for Affordable Housing

Soil stabilization is the alteration of physicomechanical properties soils to meet specific engineering requirements problematic soils. Laboratory examination well recognized as appropriate for examining stabilized soils; however, they are labor-intensive, time-consuming, and expensive. In this work, four artificial intelligence based models (OMC-EM, MDD-EM, UCS-EM+, UCS-EM−) predict optimum moisture content (OMC), maximum dry density (MDD), unconfined compressive strength (UCS) developed. Experimental data covering a wide range were collected from previously published works. The OMC-EM, UCS-EM− employed seven features that describe proportion types soils, Atterberg limits, classification groups UCS-EM+ model, besides features, employs two more describing compaction (OMC MDD). An optimizable ensemble method used fit data. model evaluation confirms developed three UCS-EM+) perform reasonably well. weak performance validates OMC MDD have substantial significance in predicting UCS. comparison all with neural network ones confirmed prediction superiority models.