This study aims to evaluate the effectiveness of sewage sludge ash (SSA) as a sustainable stabilizing agent for low-plasticity clay and to assess the applicability of machine learning techniques for predicting strength parameters. SSA was mixed with CL-type clay at contents of 0%, 5%, 10%, 15%, 20%, and 30% by dry weight. A comprehensive laboratory testing program was conducted, including Atterberg limits, compaction, fall cone, vane shear, and unconfined compressive strength (UCS) tests. The results showed that Atterberg limits increased with increasing SSA contents. As the SSA content increased up to 30%, the maximum dry unit weight decreased by approximately 6%, while the optimum water content increased by about 14%. The addition of SSA significantly enhanced the shear strength, with UCS-derived strength increments ranging from 264 to 771 kPa for 4-day-cured specimens and from 515 to 1351 kPa for 32-day-cured specimens. These findings demonstrate the strong potential of SSA as an alternative and sustainable soil stabilizer. In addition, Self-Organizing Maps (SOMs) were employed to analyze the distribution and relationships of parameters within the experimental dataset, followed by Support Vector Machines (SVMs) for classification and prediction. Using the experimental results, the undrained shear strength parameters obtained from fall cone (su-FCT) and laboratory vane shear (su-LVT) tests were predicted with accuracies of 83% and 79%, respectively. The novelty of this study lies in the integrated experimental–data-driven framework, in which extensive laboratory testing is combined with machine learning methods to both validate and reliably predict the strength behavior of SSA-stabilized clay.
Çabalar et al. (Tue,) studied this question.