Overview

Cyclistic is a substantial bike-share program that has multiple bike stations around the greater Chicago area. As the number of users grow, the company wants to explore how they can increase the number of annual memberships. Specifically, we will focus on the behaviors of casual members to create recommendations that will encourage their transition into an annual member.

Processing and Cleaning the Data

The data was gathered from a public source and made available by Motivate International Inc. found on the AWS cloud platform.

Since this is public data,there is no personal info containing the riders bank details or home address.Each ride has a unique ride ID, however, the type of membership is shown to continue our analysis.

Prior to importing the data sets the following steps were done on Excel:

  1. Calculated the ride duration for each ride in “hh:mm:ss” format
  2. Added a column indicating the day of the week in numerical format

Using R for data cleaning

The following was done with the imported csv files from Excel:

  • The ride_length column was converted to a numeric double indicating seconds
  • All NA’s were omitted from each monthly dataset

The data includes rows where the station name or end staion names are missing, but this will be analyzed later.

The monthly data frames were combined together to create a data frame observing the entire year of 2022. This data frame will be used for most of the analysis.

New columns were added to the year data set including the year, month, and date for each bike ride.

Analysis

Monthly Data

The following graph shows that there are less riders in the Winter months compared to Spring and Summer. This observation can be seen for both casual and annual members.

Daily Data

The following graph shows that casual members rode more on the weekends compared to the weekdays. Annual members appeared to rode more on the weekdays compared to the the weekends. However, overall we see that annual riders bike more consistently across the whole week compared to casual riders.

Average Ride Length: Casual VS Member

Compute the average ride length for both casual and annual members. The plots below show the average ride length for the year, month, and day of the week. The pattern we find is that the average ride length across the whole year, month, and days of the week. The average ride length of the year for casual riders was 23.84 minutes and annual riders was 12.45 minutes.

## # A tibble: 2 x 2
##   member_casual avg_ride_time
##   <chr>                 <dbl>
## 1 casual                 23.8
## 2 member                 12.5

Electric Vs Classic Bikes

We investigate the types of bikes ridden by casual and annual members across the year and each day. We find that annual riders significantly use classic bikes more than electric bikes while casual riders used around the same amount of electric and classic bikes the entire year. However, we find that the casual riders use classic and electric bikes more on the weekends. This reflects the general behavior of casual riders riding bikes more on the weekends.

Tableau Visualization

You can click here to observe the dashboard: Chicago Bike Routes and Stations

Conclusion

Based on the analysis on the data of bike rides in 2022, we found the following:

We can draw the conclusion that although annual members ride more each day we, find that casual riders use the bikes for longer periods, specifically, weekends. In addition, the Spring and Summer seasons are where most bike rides are found.Furthermore, from observations of popular bike routes, casual riders are attracted more to tourist locations like Millennium park, etc. Also, there was no clear indication of what kinds of bikes casual members preferred (electric or classic).

We can assume that there might be a larger frequency of member bike rides because it can be used more for commuting to work from Monday through Friday. We also take account the each ride ID is different but we do not know if it is the same individual riding different times each day. However, we only wanted to observe the frequency of bike rides according to type of membership.

In conclusion, some recommendations may include the following: