# Microscopic Traffic Simulation

A real-time OD demand calibration framework couples a data-driven demand-prediction network with a black-box microscopic traffic simulator in both directions, using an analytically tractable metamodel to carry gradients back so network parameters and demand estimates update together at each time step. In a 3×3 grid network, after link B1C1 closed at hour 4 on day 10, density MSE was 21.6 [veh/km]² versus 57.9 [veh/km]² for LimEKF, a 63% reduction; on the Tokyo expressway corridor during the 2021 Games travel-demand-management day, the online-updated version reached 0.66 [veh/km]² against 1.15 [veh/km]² for LimEKF.

A real-time OD demand calibration framework couples a data-driven demand-prediction network with a black-box microscopic traffic simulator in both directions, using an analytically tractable metamodel to carry gradients back so network parameters and demand estimates update together at each time step. In a 3×3 grid network, after link B1C1 closed at hour 4 on day 10, density MSE was 21.6 [veh/km]² versus 57.9 [veh/km]² for LimEKF, a 63% reduction; on the Tokyo expressway corridor during the 2021 Games travel-demand-management day, the online-updated version reached 0.66 [veh/km]² against 1.15 [veh/km]² for LimEKF.

The study presents a framework combining diffusion convolutional recurrent neural network (DCRNN) traffic and charging-demand forecasts with a CAV rerouting policy, evaluated in SUMO under controlled disruptions at five CAV penetration levels from 5% to 45%; DCRNN achieves the lowest average travel-time index, highest average speed, and lowest aggregate CO2 emissions among active routing methods, reduces average TTI by about 1.8% versus the no-reroute reference, and accepts only 41 route changes across all penetration levels versus 146 for K-shortest-path and 153 for V2X.

The study presents a framework combining diffusion convolutional recurrent neural network (DCRNN) traffic and charging-demand forecasts with a CAV rerouting policy, evaluated in SUMO under controlled disruptions at five CAV penetration levels from 5% to 45%; DCRNN achieves the lowest average travel-time index, highest average speed, and lowest aggregate CO2 emissions among active routing methods, reduces average TTI by about 1.8% versus the no-reroute reference, and accepts only 41 route changes across all penetration levels versus 146 for K-shortest-path and 153 for V2X.