With continuous increasing of mining activities, some problems, such as environmental issues, occupy a lot of space, and the risks caused by the instability of mine waste depots are far occurred than ever. One possible way to reduce mentioned problems is to stabilize and reuse mine wastes as road construction materials. On the other hand, the most significant parameter for pavement design, either using empirical or mechanistic-empirical methods, is the resilient modulus (Mr) of road materials, which shows the influence of repetitive loading on the stress-strain behavior of materials. To obtain iron ores, it is required to remove the soil resting on the iron ore storage in deeper layers. This soil is typically in the form of alluvium and is known as mine overburdens (MOs). In this study, after identification of the geotechnical characteristics of two types of MO of the Golgohar mine in Sirjan, Iran, these materials were stabilized by using three different percentages of Portland cement (5, 7, and 9%) and were cured for 7 and 28 days, respectively and the resilient modulus were measured using repetitive triaxial loading equipment at different stress levels. Results show that cement stabilization does not enhance the Mr significantly when bulk stress or confining pressure is low. As the bulk stress or confining pressure increase, the Mr of cement-stabilized MOs increases significantly compared to raw MOs. Another justification is that the Mr of cement-stabilized MOs is a function of bulk stress, and deviatoric stress has a negligible effect on the Mr. The comparison between different nonlinear models revealed that the 'Universal' model has the best fit with the measured Mr values of raw and stabilized MOs.

The resilient modulus (MR) of different pavement materials is one of the most important input parameters for the mechanistic-empirical pavement design approach. The dynamic triaxial test is the most often used method for evaluating the MR, although it is expensive, time-consuming, and requires specialized lab facilities. The purpose of this study is to establish a new model based on Long Short-Term Memory (LSTM) networks for predicting the MR of stabilized base materials with various additives during wet-dry cycles (WDC). A laboratory dataset of 704 records has been used using input parameters, including WDC, ratio of calcium oxide to silica, alumina, and ferric oxide compound, Maximum dry density to the optimal moisture content ratio (DMR), deviator stress (sigma d), and confining stress (sigma 3). The results demonstrate that the LSTM technique is very accurate, with coefficients of determination of 0.995 and 0.980 for the training and testing datasets, respectively. The LSTM model outperforms other developed models, such as support vector regression and least squares approaches, in the literature. A sensitivity analysis study has determined that the DMR parameter is the most significant factor, while the sigma d parameter is the least significant factor in predicting the MR of the stabilized base material under WDC. Furthermore, the SHapley Additive exPlanations approach is employed to elucidate the optimal model and examine the impact of its features on the final result.