Skip to content

Data Analysis | FordGoBike and Lyft Bay Wheels Bike Sharing | Descriptive Statistics

Notifications You must be signed in to change notification settings

Alexander0711/DAND_Data_Analysis_on_Lyft_FordGoBike_DataSet

Repository files navigation

DAND_Data_Analysis_on_Lyft_FordGoBike_DataSet

Exploration of FordGoBike and BayWheels data from Lyft in San Francisco to elaborate growth and business oportunities

by Alex (DAND 2020)

Dataset

For the last project of the Udacity Data Analyst Nano Degree (DAND) program I will work on the FordGoBike / Baywheels dataset.

FordGoBike and Baywheels is a public bicycle sharing company in the San Francisco Bay Area, California.

The company share a lot of their user data through a specific Data License Agreement (https://baywheels-assets.s3.amazonaws.com/data-license-agreement.html).

FordGoBike/ Baywheels dataset changed a little during those years, but here you'll find data available through all the datasets :

  • duration_sec
  • start_time
  • end_time
  • start_station_id
  • start_station_name
  • start_station_latitude
  • start_station_longitude
  • end_station_id
  • end_station_name
  • end_station_latitude
  • end_station_longitude
  • bike_id
  • user_type
  • bike_share_for_all_trip
  • rental_access_method

All data was stored in zipped CSV files. I had to download and unzip the files to reach the csvs.

Some of the data was incomplete or needed to be cleaned in more specific form.

Technical Stack

Python in Jupyter notebook by anaconda. The following librairies are used for this project in a Jupyter notebook:

  • pandas
  • numpy
  • matplotlib
  • seaborn
  • datetime
  • requests
  • os
  • io import BytesIO
  • zipfile import ZipFile
  • math
  • calendar
  • glob

Summary of Findings

  • Subscribers are doing most of the rides compared to Customers.

  • Customers are doing longer rides compared to subscribers.

  • In summer the subscribers and customers have done the most rides with peaks during March and October for 2018 and 2019.

  • The bike sharing service seems to be used by subscribers to go to and back from work and also maybe from students to go to and back from university. Therefore subscriber use the bike sharing service mainly between Monday and Friday.

  • At 8AM and 5PM is recorded the most rides of the day for subscribers and customers. But for subscribers with higher numbers.

  • Most rides have a duration between 2 to 12 minute

  • Potential for growth and gaining profit is for example to raise the bookings during the times between the rush hours by placing special offers.

Key Insights

  • Please see the PDF file "Explanatory_visualization_D01_AGO" for the key insights and visualizations

About

Data Analysis | FordGoBike and Lyft Bay Wheels Bike Sharing | Descriptive Statistics

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published