This application provides an analysis and visualization of CitiBike Usage in the Jersey City Area.

• Tableau
• Python for Data Munging

• GitHub: https://github.com/daddyjab/CitiBike
• Data Munging (Jupyter Notebook): https://github.com/daddyjab/CitiBike/blob/master/Citibike_Exploration_JAB.ipynb
• Visualization (on Tableau Public): https://public.tableau.com/profile/jeffery.brown#!/vizhome/CitiBike_with_Paths_Idle_Time_JAB_v5/StoryCitiBike
• Tableau file (tbwx): https://github.com/daddyjab/CitiBike/blob/master/CitiBike_with_Paths_Idle_Time_JAB_v5.twbx

• Jeffery Brown: Designed and implemented all code and visualations for this application.
• Data:
  • Citi Bike System Data: https://www.citibikenyc.com/system-data

This application provides a visualization of CitiBike usage data for Jersey City, New Jersey for the month of February 2019. Python and Jupyter Notebook were used to prepare the data for use in Tableau. The Tableau Deskop applicable was used to create visualization worksheets, dashboards, and a story, which were then uploaded to Tableau Public.

The Jupyter Notebook shown in Figure 1 below used used to:

• In [4]: Define a function gen_paths_df that

  • Accepts a Pandas DataFrame containing usage data, sorts the data, and generates "paths" that allow Tableau to plot the movement of bikes from one station to the next
  • Calculates useful metrics such as trip straight-line distance based upon the station coordinates, trip duration in minutes (vs. seconds), trip speed, and user age.
  • It then relabels the gender and user type fields to more meaningful terms and creates an output DataFrame with meaningful column labels.

• In [5]: Loop through each data file name to read the CSV file into a Pandas DataFrame, process the data using the gen_paths_df function, and then write the resulting DataFrame to an output CSV file.

• Even though subsequent data analysis was only performed for February 2019, this data munging was confirmed to work with several of the other larger datasets

Figure 1: CitiBike - Data Munging - Jupyter Notebook Citibike_Exploration_JAB

After initial exploration of the data in Tableau, it appeared that having data about how long bikes waited at stations before being used might be helpful. The Jupyter Notebook shown in Figure 2 below was used to generate this wait time information from the CSV files output from the previous Jupyter notebook.

Figure 2: CitiBike - Adding Idle Times - Jupyter Notebook Citibike_Exploration_Wait_Time_JAB

A number of Visualizations were created using the Tableau Desktop application and can be seen in the Tableau file CitiBike_with_Paths_Idle_Time_JAB_v5.twbx. A subset of these were subsequently arranged into a Tableau Story and uploaded to Tableau Public.

Figure 3 below shows a Tableau interactive Dashboard using Tableau Desktop on a large screen monitor.

Figure 3: CitiBike - Screenshot of Interactive Dashboard Visualization using Tableau Desktop

Figure 4 below shows a Tableau Story as viewed on Tableau Public.
Note: This visualization was designed with the smaller portrait-mode screen size supported by Tableau Public.

Figure 4: CitiBike - Interactive Story Visualization using Tableau Public