Aggregate Time Series Analysis of Urban Bus Transportation Demand in São Paulo City

Abstract

This study aims to analyze the impact of the COVID-19 pandemic on urban bus transportation demand in São Paulo, Brazil. Using time series data on passenger ridership from 2005 to 2023, we divided the dataset into two periods: pre-pandemic (2005-2020) for model generation and post-pandemic (2020-2023) for comparison of forecasted and observed values. We applied the Seasonal Autoregressive Integrated Moving Average (SARIMA) method using Jamovi software and the R programming language to analyze the time series and generate a forecasting model. Our results indicate an average forecast of 206 million passengers per month in 2023, compared to an observed average of 173 million passengers per month—a 19% shortfall. Furthermore, this forecast remains 21% below the 2019 average (220 million passengers per month). The annual projected decline between 2011 and 2023 is approximately 15.8%, resulting in a loss of over 35 million passengers annually. Notably, the decline in passenger volume predates the pandemic, which exacerbated the situation. Recovery prospects remain uncertain and hinge on factors such as sectoral investment realignment and incentives for urban bus system usage.