AI Base-Mapping of Lusaka Speeds Up Zambia's National Land Titling Programme in Informal Settlements

Zambia's National Land Titling Programme has issued only around 142,000 titled parcels across a country of 752,614 km², leaving an estimated 80% of land untitled — a gap that is sharpest in Lusaka, where roughly 70% of residents live in informal settlements with no reliable, up-to-date base map (the city's official map dated back to the 1980s). In 2021, UK Aid-funded technical agency Ordnance Survey (OS) partnered with Zambia's Ministry of Lands and Natural Resources, Ministry of Local Government, the Zambia Survey Department, the International Growth Centre (IGC) and the Commonwealth Association of Architects to build a new digital base map of a 420 km² area covering Lusaka.

OS trained a machine-learning image-classification system to identify and vectorise features from aerial imagery — roofed structures, natural surface, sealed surface, roads, grass, trees and water — producing a map that OS says was generated 'in a tenth of the time that previous traditional manual techniques would have taken.' No independent accuracy figure for the classification itself is published in the sources reviewed; the claimed gain is a production-time comparison stated by the implementing agency rather than a third-party evaluation.

Zambia's Ministry of Lands and Natural Resources has used the resulting base map to build thematic maps for national land audits, to identify land-use types within informal settlements for the titling programme, and to inform urban housing policy, climate-adaptation planning and utility infrastructure planning, according to Ordnance Survey and the International Growth Centre, an Oxford- and LSE-affiliated research institution that co-published an account of the project. OS and IGC both describe the underlying technique as replicable in other African cities facing similar informal-settlement mapping gaps, though no subsequent country has yet been documented rolling it out at the same scale.

Read the full analysis: https://www.ordnancesurvey.co.uk/insights/informing-nationwide-development-through-new-ai-land-mapping