Urban Heat Island (UHI) intensification has emerged as a pressing environmental issue in rapidly urbanizing industrial cities, particularly in developing countries like India. This study investigates the spatiotemporal dynamics of UHI in Ludhiana, a major industrial city in Punjab state of India, using an integrated geospatial and machine learning framework. The Landsat data of 1994 and 2024 were used to assess changes in built-up density and land surface temperature (LST). Urban density was mapped using supervised classification by incorporating Support Vector Machine (SVM) technique, while LST-based UHI analysis, and Getis-Ord \(G_{i}^{*}\) hotspot statistics were conducted to evaluate the thermal transformation of the urban landscape. Results reveal a substantial increase in built-up density across all urban zones (> 20%), with high-density zones (1–3) exceeding 80% coverage in 2024. Correspondingly, mean LST increased by 0.55 °C to 2.52 °C across different zones, with thermal hotspots shifting from core urban areas to peripheral and industrial zones. The results indicate that natural cooling areas such as forests and water bodies experienced significant decline, contributing to intensified heat exposure. Hotspot analysis shows a reduction in cold spot zones and an increase in emerging hot spot areas, reinforcing evidence of spatial thermal imbalance. The findings emphasize the urgent need for climate-responsive urban planning in Ludhiana, especially zones 4–8. Implementing green infrastructure, regulating industrial sprawl, and leveraging predictive spatial models can support more resilient, equitable, and thermally livable urban environments.