landslide occurrences within the high and very high susceptibility classes, significantly outperforming the Macro-Regional (37.25%) and Regional (48.26%) models. Qualitative spatial analyses further confirmed the robustness of the ANN framework, revealing a physically coherent and geomorphologically consistent distribution of susceptibility classes that closely matches documented landslide patterns across the Peruvian territory. These results demonstrate the effectiveness of Artificial Neural Networks for national-scale susceptibility mapping in complex mountainous terrains. By capturing non-linear interactions among morphometric and hydrological predictors, the proposed framework overcomes the limitations of heuristic weighting approaches while maintaining methodological simplicity and computational stability. Although developed at national scale, the methodology is inherently adaptable to multiple spatial resolutions, including mining, transport, and infrastructure applications. Beyond hazard assessment, the ANN framework provides direct operational value for the mining sector by enabling early identification of high-susceptibility zones affecting haul roads, waste dumps, tailings storage facilities, and access platforms. Its resolution-adaptive behavior supports proactive geotechnical planning, optimized infrastructure layout, and reduced disruption risk, contributing to lower operational delays and improved economic resilience. Aligned with the World Mining Congress 2026 theme, this study illustrates how advanced machine learning can support faster, smarter, and more responsible mineral production. By transforming large-scale geospatial datasets into actionable risk intelligence, the framework enhances decision-making efficiency, reduces subjectivity, and enables scalable deployment from national planning to mine-site operations without retraining. Through the integration of technological innovation and geotechnical risk management, this approach contributes to advancing transformation in the mining sector while strengthening societal trust through safer and more sustainable practices. ACKNOWLEDGEMENTS The authors acknowledge the institutional support of WSP Peru. Special thanks are extended to Martha Ly, ESG Mining Regional Leader and Earth & Environment Sector Leader for LAC, for her strategic leadership and sponsorship support. We also thank Dani Gutiérrez, Senior Environmental Specialist and Water Resources Leader at WSP Peru, for his early institutional backing in 2024, when AI-driven environmental platforms and automated systems were still emerging initiatives. His belief in the strategic potential of Digital Innovation and Geospatial Intelligence was instrumental in consolidating this function as a growing area with long-term organizational impact. Their commitment to advancing environmental excellence, technological transformation, and scientific innovation within the mining sector is deeply appreciated. REFERENCES
RkJQdWJsaXNoZXIy MTM0Mzk2