Prediction of Traffic Congestion Propagation
Modeling and quantifying time-lagged accident-induced congestion using causal inference and bootstrap uncertainty analysis.
2021.01 – 2024.02
·Motivation
- Congestion caused by an accident propagates to subsequent roads, delayed and manifested on some of them.
- Unpredictable delayed events and a lack of historical irregular-event data make the propagation pattern difficult to model.
·Goal
Identify and quantify the propagation mechanisms and time-lag effects of accident-induced non-recurrent congestion.
·Methodology
- Model the propagation pattern of non-recurrent traffic congestion.
- Identify the statistical causal relationship between accident and subsequent roads, using the bootstrap to quantify uncertainty in propagation lags.
