Lagos, Nigeria, serves as a case study for placing vertiports using geospatial machine learning. The method starts with population data as a proxy for travel demand, then layers accessibility scores based on proximity to roads and transit hubs. 100 proposed sites are generated via K-means clustering, filtered by exclusion rules targeting airports, wetlands, and water. The process highlights gaps between population density and viable vertiport locations, emphasizing the need for explicit constraints rather than algorithmic assumptions.

The approach prioritizes weighted demand—prioritizing areas with higher populations or better transport links. It replaces raw population counts with a nuanced model that accounts for real-world barriers. Murtala Muhammed International Airport's safety buffer and military zones are illustrative exclusion layers, not legal boundaries. This first-pass model is reproducible, allowing users to adapt it to other cities with similar data gaps.

Key steps include assembling public datasets (boundaries, roads, population rasters), scoring accessibility, applying hard exclusions, and reranking sites. The output is a ranked shortlist, not a final plan, acknowledging limitations like unmeasured travel patterns. The method’s strength lies in its transparency: users can audit each decision point, from data sourcing to exclusion criteria.

The Lagos example demonstrates how incomplete data can be transformed into actionable insights. By combining spatial analysis with hard rules, the model avoids overestimating viability. This framework balances technical rigor with practical adaptability, offering a template for vertiport planning in data-scarce environments.