This file is the analysis done on the bicycle data . The buisness task is the Growth of an Virtual bicycle company.

Analysis :

I have deduce the average ride duration per week day.
The number of bike rides on each week day.
How each member type use bike differnetly and their growth in number.

Installing packages

library(ggplot2)
library(dplyr)
library(tidyverse)
library(readr)
Divvy_Trips_2019_Q1_Sheet5_1_ <- read_csv("C:/project_01/Manipulating_data/findings data/Rides per Day/Divvy_Trips_2019_Q1 - Sheet5 (1).csv")
View(Divvy_Trips_2019_Q1_Sheet5_1_)


library(readr)
Divvy_Trips_2020_Q1_Sheet2_1_ <- read_csv("C:/project_01/Manipulating_data/findings data/Rides per Day/Divvy_Trips_2020_Q1 - Sheet2 (1).csv")
View(Divvy_Trips_2020_Q1_Sheet2_1_)

Glimpse of the Data

head(Divvy_Trips_2019_Q1_Sheet5_1_)

## # A tibble: 6 × 4
##   Weekdays  Rides Customer Subscriber
##   <chr>     <dbl>    <dbl>      <dbl>
## 1 Monday    50850     2556      34628
## 2 Tuesday   66778     3323      52050
## 3 Wednesday 74960     3790      57446
## 4 Thursday  69911     3348      53539
## 5 Friday    66140     3183      51751
## 6 Saturday  60663     2887      48355

Visualization

theme_set(theme_bw())
Divvy_Trips_2019_Q1_Sheet5_1_%>%
  ggplot(data = Divvy_Trips_2019_Q1_Sheet5_1_ , mapping = aes(x = Weekdays , y = Rides , fill = Weekdays))+
  geom_col( alpha = 0.5)+
  scale_x_discrete(breaks = c("Friday","Monday","Saturday","Sunday","Thursday","Tuesday","Wednesday"),
                   labels = c("Friday","Monday","Saturday","Sunday","Thursday","Tuesday","Wednesday"))+
  scale_y_continuous(breaks = c(20000,40000,60000,80000),
                   labels = c("20k","40k","60k","80k"))+
  labs(title = "No. of Rides / Day",
       x = "Weekdays ",
       y = " No. of Rides ",
       Color = "Week days")

key takeaway

The above graph shows the number of rides per each day.
There were least number of rides on Sunday.
Maximum rides were seen on Wednesday.

Adding the user factor

ggplot(Divvy_Trips_2019_Q1_Sheet5_1_ , mapping = aes(x = Weekdays , y = Subscriber , fill = Weekdays))+
  geom_col(alpha = 0.5)+
  geom_col(aes (x = Weekdays , y = Customer ) )+
  scale_x_discrete(breaks = c("Friday","Monday","Saturday","Sunday","Thursday","Tuesday","Wednesday"),
                   labels = c("Friday","Monday","Saturday","Sunday","Thursday","Tuesday","Wednesday"))+
  scale_y_continuous(breaks = c(20000,40000,60000,80000),
                     labels = c("20k","40k","60k","80k"))+
  labs(title = "Each member Rides / Day",
       x = "Weekdays ",
       y = " Subscriber ",
       Color = "Week days")

key takeaway * There denser area shows the Daily customer rides. * The light area shows the daily Subscriber rides. * Subscribers uses more bikes often than customer.

Difference in bike ride duration

Divvy_Trips_2020_Q1_Sheet2_1_ %>%
  drop_na()%>%
  group_by(`Week days` , `Duration customer 2019` , `Duration subscriber 2019` , `Duration Customer 2020`,`Duration Subscriber 2020`)

## # A tibble: 7 × 8
## # Groups:   Week days, Duration customer 2019, Duration subscriber 2019,
## #   Duration Customer 2020, Duration Subscriber 2020 [7]
##   `Week days` Customer Subscriber Rides `Duration Customer 2020`
##   <chr>          <dbl>      <dbl> <dbl> <chr>                   
## 1 Monday          8866      41984 50850 13302:48:14             
## 2 Tuesday         6576      60202 66778 9495:07:28              
## 3 Wednesday       7796      67164 74960 12587:15:45             
## 4 Thursday        7763      62148 69911 7831:11:40              
## 5 Friday          6569      59571 66140 12714:20:40             
## 6 Saturday        6089      54574 60663 10292:14:22             
## 7 Sunday          4821      32764 37585 11169:24:26             
## # ℹ 3 more variables: `Duration Subscriber 2020` <chr>,
## #   `Duration customer 2019` <chr>, `Duration subscriber 2019` <chr>

week_days <- c("Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday")
customer_duration <- c(798168.2, 569707.5, 755235.8, 469871.7, 762860.7, 617534.4, 670164.4)
subscriber_duration <- c(708645.9, 734922.7, 800916.3, 733588.5, 699680.4, 676067.1, 443287.3)

# Convert to a data frame
data <- data.frame(Weekday = factor(week_days, 
                                    levels = c("Monday", "Tuesday", "Wednesday", "Thursday", 
                                               "Friday", "Saturday", "Sunday")),
                   Customer = customer_duration, 
                   Subscriber = subscriber_duration)

ggplot(data, aes(x = Weekday, group = 1)) +  # Ensure group aesthetics are set
  geom_line(aes(y = Customer, color = "Customer"), size = 1) +
  geom_line(aes(y = Subscriber, color = "Subscriber"), size = 1) +
  scale_y_continuous(breaks = c(554.2793056,
                                395.6302083,
                                524.4693056,
                                326.2997917,
                                529.764375,
                                428.8433333,
                                465.3919444),
                     labels = c("554hr","395hr","524hr","326hr","529hr","428hr","465hr"))+
  labs(title = "Duration of Rides in 2020",
       x = "Week Days", 
       y = "Duration (Minutes)",
       color = "User type") +
  scale_color_manual(values = c("Customer" = "blue", "Subscriber" = "red")) +
  theme_minimal()

Conclusion:

There were more number of bike rides by subscribers.
There was a significant growth in subscriber in 2020.
Members use bike more often than Customers.

Bicycle_Analysis

Binny