The caving and subsidence developments above a longwall panel usually result in fractures of the overburden, which decrease the strength of the rock mass and its function. The height of fracturing (HoF) includes the caved and continuous fractured zones affected by a high degree of bending. Among the various empirical models, Ditton’s geometry and geology models are widely used in Australian coalfields. The application of genetic programming (GP) and gene expression programming (GEP) in longwall mining is entirely new and original. This work uses a GEP method in order to predict HoF. The model variables, including the panel width (W), cover depth (H), mining height (T), unit thickness (t), and its distance from the extracted seam (y), are selected via the dimensional analysis and Buckingham’s P-theorem. A dataset involving 31 longwall panels is used to present a new nonlinear regression function. The statistical estimators, including the coefficient of determination (R2), the average error (AE), the mean absolute percentage error (MAPE), and the root mean square error (RMSE), are used to compare the performance of the discussed models. The R2 value for the GEP model (99%) is considerably higher than the corresponding values of Ditton’s geometry (61%) and geology (81%) models. Moreover, the maximum values of the statistical error estimators (AE, MAPE, and RMSE) for the GEP model are 12%, 14%, and 16%, respectively, of the corresponding values of Ditton’s models.

Accurate prediction of surface subsidence becomes a significant challenge for active industrial companies in coal mining fields due to the importance of the economic impacts of longwall mining-induced subsidence. This article explores a new variant of genetic programming, namely gene expression programming (GEP). The GEP-based method is utilized to present a new mathematical formula for subsidence prediction in longwall coal mining. The derived model includes both geometrical and geological variables. The data set consists of field measurements obtained through 37 longwall panels of Ulan coal mine, NSW, Australia. The GEP-based model concluded satisfactory subsidence prediction outcomes compared to other empirical methods such as NCB, DMR, ACARP, and IPM. The predictive ability of the GEP-based models, which captures the complex nonlinear effects of the critical factors on the magnitude of subsidence, resulted in a statistically significant improvement in predictive capacity compared to the aforementioned empirical methods. The sensitivity analysis results indicated that Panel width and cover thickness with 31% and 23% were the most influential parameters in the proposed model. Also, the extracted seam thickness, thick layer location, and thick layer thickness had 19%, 16%, and 11% impact on the GEP proposed model, respectively.