Bike_Sharing

Purpose

Client has requested Tableau visualizations of NYC CitiBike bikeshare trip data to pitch to investors on a similar product for Des Moines, Iowa.

Resources:

• Tableau Public
• NYC CitiBike August 2019 Rideshare Data.

Deliverables

• Convert trip duration data to a datetime format with Python.
• Atleast 7 graphs
• Create a Tableau Story for presentation.

What is the idea?

Presentation

• An interactive Tableau Story Board is available to play with and filter yourself on the Tableau Public Hosting service.

• The Dashbords are able to be filtered by Usertype, Gender and Week by clicking on the designated colored square.

Summary

CitiBike is a bicycle renting service powered by Lyft with stations throughout the city of New York. I'd like to start by defining a few terms first:

• Gender is broken into Male, Female, and Unknown because in addition to the people with non-binary genders that make up most of the category, there are people that refuse to answer the question.
• CitiBike classifies their users into one of 2 categories: Annual Subscribers (Subscriber) and Everyone else (Customer). Subscribers are mostly commuters while Customers are casual riders, tourists, and new riders.

• With an inventory of 24,00+ bicycles and 1,500+ stations, 13,983 bicycles and 807 bike stations were active during the month of August 2019.
• 2,344,224 Rides dring the month of August 2019 with a total ride time of 41,030,675 minutes, equivalent to 78 years if you were to measure it as one continuous stream.

• The most common riders during the month of Augaust 2019 were male subscribers during "rush hour" for an average ride time of 15 to 20 minutes.
• The best time of day to shuffle bikes around optimizing utilization will be Midnight to 5am.
• There are stations in 4 of the 5 buroughs but the ones in Manhattan are the most popular.

• Bicycle maintenece will need performed most on the bicycles with a high ride count or High total ride time. Retrieving those bikes can be accomplished as apart of the nightly rotation of bicylces that need to change stations for best utilization.

Recommended Improvements to the Analysis

• Add the ability to filter the stations by burough/neighborhood.
• Find the typical radius a bicycle travels, do they tend to stay in their initial neighborhoods or traverse the borders?

Additional Data required to:

• Compare station locations with bus and train stops/routes.
• Compare station location and popularity to surrounding business types.
• Compare cost of living and incomes to ridership in neighborhoods and stations.
• Use more than one month, use the data for a financial quarter or a year. How does this affect the analysis?

References:

• Ride Sharing Data from CitiBike's historical data (August 2019), which they store here.