Landslides pose a significant threat to infrastructure and human life along the National Highway 10 (NH10) corridor in the Indian Himalayas. This study employed cutting-edge spatial modeling techniques—Analytical Hierarchy Process, Frequency Ratio, and Random Forest—to map landslide susceptibility and inform risk mitigation efforts. A comprehensive geo-database was created incorporating 11 conditioning factors, including slope, elevation, rainfall, distance to rivers, distance to roads, distance to lineament, topographic wetness index, normalized difference vegetation index, land use, geology, and aspect. A total of 304 historical landslide locations were identified and randomly divided into 70% training and 30% validation datasets. The resulting landslide susceptibility map categorizes the road into zones of varying landslide susceptibility levels, ranging from very low to very high. Approximately 33%, 44.64% and 47.51% of the road falls within the high and very high susceptibility categories according to the Analytical Hierarchy Process, Frequency Ratio, and Random Forest methods respectively. The susceptibility maps generated by the three methods were validated using the area under the receiver operating characteristic (ROC) curve and the success rate curve. The results showed that the RF model outperformed the AHP and FR models, with an area under the ROC curve of 0.859 for the validation dataset. The finer granularity and more conservative high-risk delineation of the RF model suggest it may be the most suitable approach for targeted risk mitigation planning and infrastructure development along the critical NH10 corridor in the Indian Himalayas. These findings can empower data-driven decision-making to safeguard lives, livelihoods, and transportation links in this treacherous region.