Forecasting Interborough Express Ridership Using Network-Based Station Typologies and Direct Demand Models

Forecasting ridership for new transit infrastructure is difficult in the absence of observed outcomes, particularly under domain shift between an existing system and a proposed corridor. This study develops a station-level direct demand modeling (DDM) framework to forecast average weekday ridership for the proposed Interborough Express (IBX) in New York City — a 14-mile circumferential rapid transit corridor connecting Brooklyn and Queens. The approach pairs unsupervised learning with supervised estimation in a common feature space defined by transit service, accessibility, and built-environment characteristics. K-means clustering identifies latent station typologies (node–place regimes), and IBX stations are projected into this topology to define a corridor-relevant training subset, a strategy termed cluster-based domain focusing. Forecasting models — ordinary least squares, regularized linear estimators (Ridge, Lasso, Elastic Net), and tree-based ensembles (Random Forest, Gradient Boosting, XGBoost) — are evaluated through holdout testing and cross-validation. Regularized linear models outperform nonlinear alternatives in both predictive accuracy and stability, with Ridge regression selected as the preferred specification (out-of-sample R² ≈ 0.72 in the IBX-restricted domain). The resulting baseline forecast is approximately 61,000 average weekday entries, substantially below the MTA’s official projection of 115,000. Cross-model agreement is high among regularized estimators, indicating that the dominant predictive signal is approximately linear in log-transformed space and that divergence from official projections arises primarily from scenario assumptions — induced demand, long-run growth, and full network reconfiguration — rather than model misspecification. The results demonstrate that feature-space–based domain restriction provides a principled, empirically grounded framework for forecasting transit ridership under missing outcomes.