Figure 1
From: Real-time urban traffic amount prediction models for dynamic route guidance systems
Prediction errors versus prediction interval Ï„ . (a) MAE (b) SMAPE. Model-1 has the best prediction accuracy and reduces MAE and SMAPE by up to 52% and 41% respectively compared to the Shift Model in the best case (Ï„=10 s). Model-2 also reduces MAE and SMAPE by up to 30% (Ï„=10 s) and SMAPE by 28% (Ï„=300 s) compared to the Shift Model in the best cases. Reducing Ï„ may improve prediction accuracy, but short Ï„ is not always achievable as it is bounded by the latency of traffic data collection and data processing at the control center.