In the heart of Egypt’s Nile Delta, Mansoura City is quietly becoming a testbed for how artificial intelligence can reshape urban planning—and it’s happening through an unlikely marriage of satellite images and conversational AI. A new study led by Hani Abdelbasset ElShafei from the Faculty of Arts at Mansoura University introduces a system that turns raw geospatial data into actionable insights for city officials, using deep learning and natural language processing to automate land-use analysis and answer complex planning questions in plain English.
The system begins with a high-resolution satellite image of Mansoura. Using a DeepLabV3+ model—a type of neural network commonly used in image segmentation—the AI identifies and labels different land covers: where buildings are spreading, where green spaces remain, and where water bodies lie. This isn’t just a static map; it’s a living layer of intelligence that updates as new imagery becomes available. “We’re not just mapping the city,” ElShafei explains. “We’re teaching the city to speak back in a language planners can understand.”
Once the land cover is classified, the system moves to its second phase: a conversational interface powered by Retrieval-Augmented Generation (RAG). City officials can now type questions like, “Show me the areas with the fastest urban growth in the last five years,” and receive not only a highlighted map, but a synthesized response grounded in both the spatial data and urban planning literature. The platform, built as a FastAPI web application, combines interactive maps with an AI assistant that feels less like software and more like a planning colleague who never sleeps.
For the energy sector, the implications are significant. Urban sprawl doesn’t just mean more buildings—it means longer power lines, higher energy demand peaks, and new pressure on water and drainage systems. Utilities and renewable energy developers can use this kind of real-time LULC intelligence to anticipate where new substations, solar farms, or water pipelines might be needed before bottlenecks form. Instead of reacting to growth, they can plan for it.
The study, published in *Mağallaẗ Kulliyyaẗ Al-Adāb, Ǧāmiʿaẗ Al-Zaqāzīq* (Journal of the Faculty of Arts, Zagazig University), is more than an academic exercise. It’s a blueprint for how AI can democratize geospatial analysis, putting powerful tools into the hands of non-technical decision-makers. As ElShafei notes, “The goal isn’t to replace planners—it’s to give them a faster, clearer lens to see the city as it changes.” For energy companies navigating rapid urbanization, that lens could be worth more than gold.