This study develops and applies a geospatially driven computational framework to enhance the operational efficiency of irrigation canals, demonstrated through the Kadi branch of the Narmada Main Canal in Gujarat, India. Canal seepage and subsequent waterlogging are major contributors to reduced irrigation efficiency and secondary salinization in command areas. To characterize these processes, multi-temporal Landsat datasets (1990–2024), high-resolution UAV Ortho-mosaics, and ground-based geophysical measurements were analysed to generate long-term vegetation and surface-moisture indices, specifically the Normalized Difference Vegetation Index (NDVI) and the Normalized Difference Water Index (NDWI). A multi-objective optimization model, formulated on the principles of the Non-Dominated Sorting Genetic Algorithm II (NSGA-II), was implemented to identify intervention strategies that minimize seepage losses and waterlogged area while sustaining irrigation deliveries. The analysis revealed recurring moisture persistence and vegetative anomalies adjacent to the canal alignment, confirming progressive seepage patterns. Optimization results indicated that selective lining of high-loss segments combined with targeted sub-surface drainage could achieve approximately 20% reduction in seepage without adversely affecting supply reliability. The study demonstrates how the integration of remote sensing, UAV data, and evolutionary algorithms can support data-driven, cost-effective canal management, contributing to more sustainable and resilient irrigation infrastructure planning in India.