Conclusions

Historically, Washington DC has had a congestion problem. The Covid-19 pandemic temporarily reduced the amount of traffic, but congestion is already nearing pre-pandemic numbers. This analysis analyzed current and future ridership for alternative transportation methods in the hope of better understanding how to get people out of the car. Adequate access to public transportation is one of the most important factors in an individual’s decision to take a car or take public transportation. By forecasting future ridership numbers for different modes of public transportation these systems can ensure that they will continue to provide enough service for the number of passengers. In addition better understanding cyclical trends of ridership will also help these systems maintain adequate transportation.

This analysis found that Bus and Metro ridership had been decreasing pre-pandemic and sharply fell due to the pandemic. All models agreed that both Bus and Metro ridership will continue to increase over the next 6 months, indicating that WMATA (Washington metropolitan area transit authority) should continue to increase the number and frequency of trains and buses. This coincides with what WMATA is doing, as they plan to scale up the frequency of trains through July of 2023.

This analysis also found that Capital Bikeshare usership has remained steady since 2015, and was not nearly as impacted by the pandemic. Slight increases in ridership are expected over the next year, so we recommend that Capital Bikeshare continue to invest in more bikes, and stations. Of specific interest in this analysis is that Bike Share ridership is particularly low on rainy and extreme cold days and extremely high on days in which DC is hosting big events.

This is an important finding for Capital Bikeshare as it indicates that they need to be doing more distribution of bikes during the summer months and on days with big events. Additionally, in the winter months and on rainy days investing in the redistribution of bikes likely has less impact.

Of the standard time series modelling methods, those that were able to account for multiple variables were the most successful. This indicates that Bus, Metro, and Bike ridership are all determined by other factors than just previous ridership. Further analysis should take into account other factors that may affect ridership such as holidays, the day of the week, economic factors and more.

While the findings of this analysis were important there were still high levels of uncertainty in the predictions, indicating that more data is needed to accurately forecast future ridership. Additionally, it was difficult to account for the shock that the pandemic had on transportation. As the world continues to adjust to the post pandemic normal, better forecasts will be possible.