<![CDATA[This study addresses the seasonal and long-term fluctuating passenger volume patterns typical of water transportation systems such as NYC Ferry, necessitating practical forecasting methods to support operational decision-making and public transportation planning. The research aims to develop a forecasting model for NYC Ferry passenger counts using the Seasonal Autoregressive Integrated Moving Average (SARIMA) methodology. The analysis utilizes monthly historical passenger data from January 2020 to December 2024 for training data. Key analytical steps include testing data stationarity, splitting the dataset into training and testing subsets, modeling via RStudio, forecasting, and evaluating model accuracy using Mean Absolute Percentage Error (MAPE) compared against actual observations. Results indicate that the SARIMA(1,0,0)(0,1,1)12 model outperforms other methods, yielding the lowest MAPE of 5.04%, compared to Multiplicative Winters (8.57%), SARFIMA (17.62%), and Holt-Winters (32.93%). The SARIMA model effectively captures both seasonal and monthly trends, producing accurate passenger volume predictions. These findings demonstrate SARIMA’s efficacy in monthly NYC Ferry ridership forecasting, contributing to time series literature, particularly within public transportation forecasting. Furthermore, the results offer practical insights for policymakers to strategize service capacity and enhance data-driven management of waterborne transit systems more efficiently.]]>
