Risk-Aware Planning for Transit Desert Remediation Under Demand Uncertainty
Transit deserts are areas where public transportation is inadequate despite evidence of travel demand, a condition that affects tens of millions of residents across the Americas. Planning for these areas is difficult because the usual demand signal is missing: ridership cannot be observed before service exists. To address that setting, we formulate risk-aware transit desert remediation as a partially observable Markov decision process with Conditional Value-at-Risk constraints for financial tail risk. The model uses demographic, land-use, and employment data to set a prior over latent demand, then updates that prior as new service deployments produce ridership observations. A myopic belief-aware planner is evaluated on 25 cities using a unified financial model for operating cost, capital expenditure, fare revenue, and net subsidy. After five years, the planner remediates a median of 53.6% of transit-desert tracts and improves on static optimization by 5.0 percentage points on average, with gains in 16 of 25 cities. Gains are largest at moderate budgets (+9.9 points at baseline) and persist under 50% prior-demand miscalibration, while population density and existing transit density are the strongest structural predictors of remediation cost ($R^2\!=\!0.41$ on per-tract cost)