Assessment of Rock Slope Stability Using Mobile LiDAR-Aided Hybrid Classification Systems: A Comparative Analysis of SMR, Q-Slope, and RHI

ABSTRACT This study evaluates andesite rock slope stability in İzmir, Türkiye, using mobile LiDAR (iPhone 14Pro) and hybrid classification systems. The acquired 3D point cloud (1.7 × 106 points) demonstrated high spatial accuracy when validated against global navigation satellite system data. Automated analysis via discontinuity set extractor revealed that applying a “full-persistence” assumption artificially reduces mean discontinuity spacing by 40–70 percent compared with a realistic “non-persistence” model, drastically underestimating rock block sizes. Kinematic analysis indicated a 40 percent probability of wedge failure, primarily driven by the highly persistent J6 set (lmax = 16 m). To directly compare slope mass rating (SMR), Q-slope, and rock hazard index (RHI) outputs, a “normalized risk index (Rindex)” was utilized. SMR yielded an overly optimistic evaluation Rindex = 0.41, failing to capture the geometric severity of the critical wedge. In contrast, RHI successfully identified this localized hazard (“moderate risk,” Rindex = 0.51) by directly accounting for geometric continuity. The Q-slope system proved to be the most conservative (Rindex = 0.60), indicating that the current 72° slope requires a reduction to 57°. The findings demonstrate that integrating mobile LiDAR with geometry-sensitive systems such as RHI provides a more reliable, targeted hazard assessment than traditional approaches.