Assessment of slope stability based on the improved entropy weight-gray target theory

Landslides are a serious natural hazard globally. While the traditional grey target model has long served as an effective tool for slope stability assessment, its predictive accuracy remains questionable due to the inherent correlations among evaluation indicators. To mitigate this limitation, this study introduces an enhanced entropy weight-grey target theory for slope stability evaluation. The proposed model innovates in two key aspects: first, it substitutes the covariance matrix in the Mahalanobis distance calculation with a correlation coefficient matrix, thereby addressing indicator interdependencies; second, it adapts the positive and negative ideal solutions from the TOPSIS model to define the corresponding target centers in the grey target framework. The improved model’s efficacy is validated through an engineering case study. The findings confirm that the proposed method not only offers a feasible approach for slope stability assessment but also demonstrates superior predictive accuracy to the traditional grey target model. This research contributes a novel methodology and conceptual framework for future slope stability evaluations.