NYC Citibike Trip Analysis with Tableau

Objective:

The goal of this project is to gain insight into the NYC Citibike data for 2021.

Tools & databases used:

• Jupyter
• Pandas
• NYC Citibike Data
• Tableau

Tableau analysis:

Dashboard Analysis
Story Analysis
Please visit above links to view the tableau project!

Preparing my data:

Collecting the data

I'm downloading all the 2021 data for NYC citibike which is openly attainable to the public as a csv file.

Cleaning with pandas:

To start saw that the data in the start_at and ending_at columns were dates, so I formatted them to datetime.

After, I noticed they removed the "tripduration" column. So, to start I took "ending_at" and subtracted "started_at" and divided by panda's timedelta with a parameter of seconds. Then converting that column to datetime for better analysis parameters in Tableau.

raw_nycitibike_df['duration_seconds'] = (raw_nycitibike_df.ended_at - raw_nycitibike_df.started_at) / pd.Timedelta(seconds=1)
raw_nycitibike_df['duration_seconds'] = pd.to_datetime(raw_nycitibike_df['duration_seconds'], unit='s')

Final step was to trim the dataset. The original file contains 26,500,000 rows of data, but Tableau Public only allows 15,000,000. So I trimmed off what I needed to in order to load it into Tableau.

Calculated field with tableau:

I noticed the datetime was in military format, so within tableau I created a new calculated field to convert 24 hours to 12 hour time.

IF STR(DATEPART('hour',[Started At]))='12' THEN "12 PM"
ELSEIF STR(DATEPART('hour',[Started At]))='0' THEN "12 AM"
ELSEIF DATEPART('hour',[Started At])>12 THEN STR(DATEPART('hour',[Started At])-12)+" PM"
ELSE STR(DATEPART('hour',[Started At]))+" AM"
End

This produces an unsorted result (1AM, 1PM, 2AM...), so I had to go in and manually adjust the sorting so that it displayed data correctly.