Forecasting Daily and Weekly Passenger Demand for Urban Rail Transit Stations Based on a Time Series Model Approach

Forecasting daily and weekly passenger demand is a key fundamental process used by existing urban rail transit (URT) station authorities to diagnose operational problems make decisions about train schedule patterns improve efficiency, increase revenue management, driving safety. The accuracy of the forecast results will directly affect operation planning (URT). Therefore, based on collected inbound historical data, this study Box–Jenkins time series with Facebook Prophet algorithm analyze characteristics achieved better computational forecasting performance accuracy. After analyzing periodicity, correlation, stationarity, different models were constructed. Akaike information criteria (AIC), Bayesian (BIC), mean squared error (MSE), root (RMSE) evaluate adequacy best model from among several tested candidates’ for Box–Jenkins. parameters estimated using statistical software. experimental are both theoretical practical significance an effective system. signify that SARIMA (5, 1, 3) (1, 0, 0)24 performs more stable in demand, ARMA (2, 1) demand. When comparing Prophet, show superior series, series.