RideNow uses a gradient-boosted regression model to predict next-day driver demand by city zone and hour so operations can position incentives and staffing. Over the last 6 weeks, business leaders reported that predicted demand has become materially different from actual completed rides, causing overspending in some zones and driver shortages in others.
| Metric | Validation at Launch | Current Production | Change |
|---|---|---|---|
| RMSE | 18.4 rides/hour | 31.7 rides/hour | +72.3% |
| MAE | 11.2 rides/hour | 19.6 rides/hour | +75.0% |
| MAPE | 8.9% | 16.8% | +7.9 pts |
| Bias (Predicted - Actual) | +0.8 rides/hour | +9.4 rides/hour | +8.6 |
| P90 Absolute Error | 27 rides | 49 rides | +81.5% |
| Zones with abs error > 20 rides | 14% | 37% | +23 pts |
The model is systematically overpredicting demand in suburban commuter zones and underpredicting demand near airports and event venues. Operations estimates that forecast error increased weekly incentive waste by $180K and reduced fulfilled ride rate by 2.7 percentage points during peak periods.