Disconnecting pile heads from the raft has gained wide application because the gravel and geogrid layers filling the space between them create a more even pressure distribution and reduce differential settlement. However, due to the complexity of modelling the multiple interfaces in the superstructure-foundation-subsoil system, previous findings on load-transfer mechanisms, interaction effects, and group optimization remain incomplete. Moreover, the common analytical approach employed for understanding the load transfer mechanism is the “unit cell’ concept, which cannot fully capture dissimilarity in the group. To address these limitations, this study aims to develop an analytical framework for predicting the load-settlement response of cushioned piled rafts with dissimilar piles. The proposed method simplifies the cushion-pile-soil interaction using a Winkler-type Spring model, while a hyperbolic load-transfer function captures the nonlinear pile-soil behavior. The model was verified against existing experimental and showed close agreement. It successfully captured the load-sharing mechanism, confirming that stiffer, longer, or larger-diameter piles attract a disproportionately higher share of the load. The novelty of this work lies in the establishment of an analytical model based on the principles of dissimilar pile groups but extended to include cushion force transmission, a critical integration that provides a realistic tool for practice.

Expansive soils, characterized by pronounced volumetric instability under moisture fluctuations, pose a persistent challenge to the safety, serviceability, and long-term stability of civil infrastructure. In response to these challenges, increasing attention has been directed toward the use of sustainable waste-derived additives as environmentally responsible alternatives for improving problematic soils. This study investigates the mechanical behavior of expansive soil stabilized with waste eggshell powder (ESP) and recycled tire powder (RTP), an elastic polymeric waste material rich in rubber and carbon black derived from discarded tires. A comprehensive experimental program was conducted to evaluate the combined effects of ESP/RTP on soil performance. For this purpose, Atterberg’s limits, compaction, unconfined compressive strength (UCS), Swell potential, and California bearing ratio (CBR) tests were performed on untreated soil and soil modified with up to 25% ESP and 9% RTP. The experimental findings revealed that the incorporation of ESP and RTP significantly reduced the Atterberg limits, optimum moisture content (OMC), and swelling potential of the soil, with the optimum blend of 20% ESP and 6% RTP yielding the most pronounced overall improvement. Such changes signify a considerable reduction in soil plasticity and expansive susceptibility, thereby enhancing the volumetric stability of the treated soil. At the same time, maximum dry density (MDD), UCS, and CBR values increased appreciably, demonstrating improved compaction behavior, greater mechanical resistance, and superior load-bearing performance. Moreover, one-way ANOVA confirmed that the enhancements in UCS and CBR achieved at the optimum mix proportions were statistically significant at the 95% confidence level. Overall, the findings demonstrate that ESP and RTP can serve as a sustainable, cost-effective, and environmentally friendly stabilization solution for improving the engineering performance of expansive soils.