CYCLIST BIKE SHARE

Analysis
Code
EDA
Author

Ajay Shankar A

Published

December 14, 2024

1 CYCLIST BIKE SHARE

1.1 Introduction

The analysis is done on Cyclist Trip Data obtained from Coursera Google Data Analytics course as part of Cap Stone Project.

The data contains month wise travel usage of bikes from the year of 2015-2023. We will be concentrating on data gathered in between July-2022 to June-2023 which will cover an entire year.

Let’s load the required packages first

  • Loading tidyverse and gt packages
library(tidyverse)
library(gt)

1.1.1 Loading and Formatting Data

  • Let’s look at the structure of the data in one of the downloaded .csv files.
trpdata_july_2022<-read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202207-divvy-tripdata/202207-divvy-tripdata.csv")
Rows: 823488 Columns: 13
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr  (7): ride_id, rideable_type, start_station_name, start_station_id, end_...
dbl  (4): start_lat, start_lng, end_lat, end_lng
dttm (2): started_at, ended_at

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
glimpse(trpdata_july_2022)
Rows: 823,488
Columns: 13
$ ride_id            <chr> "954144C2F67B1932", "292E027607D218B6", "5776585258…
$ rideable_type      <chr> "classic_bike", "classic_bike", "classic_bike", "cl…
$ started_at         <dttm> 2022-07-05 08:12:47, 2022-07-26 12:53:38, 2022-07-…
$ ended_at           <dttm> 2022-07-05 08:24:32, 2022-07-26 12:55:31, 2022-07-…
$ start_station_name <chr> "Ashland Ave & Blackhawk St", "Buckingham Fountain …
$ start_station_id   <chr> "13224", "15541", "15541", "15541", "TA1307000117",…
$ end_station_name   <chr> "Kingsbury St & Kinzie St", "Michigan Ave & 8th St"…
$ end_station_id     <chr> "KA1503000043", "623", "623", "TA1307000164", "TA13…
$ start_lat          <dbl> 41.90707, 41.86962, 41.86962, 41.86962, 41.89147, 4…
$ start_lng          <dbl> -87.66725, -87.62398, -87.62398, -87.62398, -87.626…
$ end_lat            <dbl> 41.88918, 41.87277, 41.87277, 41.79526, 41.93625, 4…
$ end_lng            <dbl> -87.63851, -87.62398, -87.62398, -87.59647, -87.652…
$ member_casual      <chr> "member", "casual", "casual", "casual", "member", "…
  • Let’s look at the columns and try to understand what they represent
    • ride_id is the unique identification token generated for each ride that was initiated.
    • rideable_type indicates the type of bike used for the ride.
    • started_at and ended_at give us the time when the ride began and the ride ended respectively.
    • start_station_name and end_station_name give us the names of stations where ride began and ended respectively.
    • start_station_id and end_station_id are unique ID’s given to stations.
    • start_lat and start_lng represent co-ordinates where the ride began.
    • end_lat and end_lng represent co-ordinates where the ride stopped.
    • member_casual identifies if the rider is a member or casual rider of the bike.

The trpdata_july_2022 contains 823488 rows and 13 columns. In the results we can see all the columns and their data types.

  • Lets load data of remaining 11 months.
trpdata_aug_2022 <- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202208-divvy-tripdata/202208-divvy-tripdata.csv")

trpdata_sept_2022<- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202209-divvy-tripdata/202209-divvy-publictripdata.csv")

trpdata_oct_2022<- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202210-divvy-tripdata/202210-divvy-tripdata_raw.csv")

trpdata_nov_2022<- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202211-divvy-tripdata/202211-divvy-tripdata.csv")

trpdata_dec_2022 <- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202212-divvy-tripdata/202212-divvy-tripdata.csv")

trpdata_jan_2023 <- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202301-divvy-tripdata/202301-divvy-tripdata.csv")

trpdata_feb_2023 <- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202302-divvy-tripdata/202302-divvy-tripdata.csv")

trpdata_mar_2023 <- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202303-divvy-tripdata/202303-divvy-tripdata.csv")

trpdata_apr_2023 <- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202304-divvy-tripdata/202304-divvy-tripdata.csv")

trpdata_may_2023 <- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202305-divvy-tripdata/202305-divvy-tripdata.csv")

trpdata_june_2023 <- read_csv("F:/Data_Sci/Cap_Stone_Project/Cyclist_trip_data/202306-divvy-tripdata/202306-divvy-tripdata.csv")

As structure of .csv’s is same across the all the files lets combine all the .csv files into a single data frame which contains data of all 12 months.

  • Combining all the monthly data to one previous year data(data_one_year_raw).
data_one_year_raw <- rbind(trpdata_july_2022, trpdata_aug_2022,
                     trpdata_sept_2022, trpdata_oct_2022,
                     trpdata_nov_2022, trpdata_dec_2022,
                     trpdata_jan_2023, trpdata_feb_2023,
                     trpdata_mar_2023, trpdata_apr_2023,
                     trpdata_may_2023, trpdata_june_2023)

glimpse(data_one_year_raw)
Rows: 5,779,444
Columns: 13
$ ride_id            <chr> "954144C2F67B1932", "292E027607D218B6", "5776585258…
$ rideable_type      <chr> "classic_bike", "classic_bike", "classic_bike", "cl…
$ started_at         <dttm> 2022-07-05 08:12:47, 2022-07-26 12:53:38, 2022-07-…
$ ended_at           <dttm> 2022-07-05 08:24:32, 2022-07-26 12:55:31, 2022-07-…
$ start_station_name <chr> "Ashland Ave & Blackhawk St", "Buckingham Fountain …
$ start_station_id   <chr> "13224", "15541", "15541", "15541", "TA1307000117",…
$ end_station_name   <chr> "Kingsbury St & Kinzie St", "Michigan Ave & 8th St"…
$ end_station_id     <chr> "KA1503000043", "623", "623", "TA1307000164", "TA13…
$ start_lat          <dbl> 41.90707, 41.86962, 41.86962, 41.86962, 41.89147, 4…
$ start_lng          <dbl> -87.66725, -87.62398, -87.62398, -87.62398, -87.626…
$ end_lat            <dbl> 41.88918, 41.87277, 41.87277, 41.79526, 41.93625, 4…
$ end_lng            <dbl> -87.63851, -87.62398, -87.62398, -87.59647, -87.652…
$ member_casual      <chr> "member", "casual", "casual", "casual", "member", "…
  • data_one_year_raw data frame contains data from the month of July-2022 to June-2023.

1.1.2 Cleaning the data

  • Checking and counting “NA” in each column of the data frame. Data is much better without “NA” as they can cause problems while aggregating data and calculating averages and sums. We can use map function to perform a function to all of the columns.
na_in_cols <- data_one_year_raw %>% map(is.na) %>% map(sum) %>% unlist()

na_in_cols
           ride_id      rideable_type         started_at           ended_at 
                 0                  0                  0                  0 
start_station_name   start_station_id   end_station_name     end_station_id 
            857860             857992             915655             915796 
         start_lat          start_lng            end_lat            end_lng 
                 0                  0               5795               5795 
     member_casual 
                 0

  • As NA’s are not present in the times columns i.e, started_at and ended_at we don’t need to worry ourselves about writing na.rm during aggregation and manipulation of data but it is a good practice to do so.

  • Finding the length or duration of the rides by making a new column ride_length in minutes and making sure that the ride_length is not negative by using if_else function. Eliminating stations where station names and longitude and latitude co-ordinates are not present.

# As we remove all the NA's it is better to save the data as "data_one_year".
data_one_year <- data_one_year_raw %>% 
  mutate(ride_length = difftime(ended_at, started_at,
                                units = "min")) %>%
  mutate(ride_length = as.numeric(ride_length))

data_one_year <- data_one_year %>%
  mutate(ride_length = if_else(ride_length < 0, 0, ride_length)) %>% 
  filter(ride_length >= 2,
         start_station_name != "" & end_station_name != "" & 
         !is.na(start_lat) & !is.na(start_lng) &
         !is.na(end_lat) & !is.na(end_lng)) %>% arrange(ride_length) %>% 
  select(ride_id, rideable_type, ride_length,
         started_at, ended_at, member_casual)


glimpse(data_one_year)
Rows: 4,243,652
Columns: 6
$ ride_id       <chr> "898EAA520DDCF78F", "50BACAD085808776", "961FDB38764FE54…
$ rideable_type <chr> "classic_bike", "classic_bike", "classic_bike", "classic…
$ ride_length   <dbl> 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,…
$ started_at    <dttm> 2022-07-27 17:51:57, 2022-07-31 20:02:35, 2022-07-06 18…
$ ended_at      <dttm> 2022-07-27 17:53:57, 2022-07-31 20:04:35, 2022-07-06 18…
$ member_casual <chr> "member", "member", "casual", "member", "casual", "membe…

1.2 Analysis of Data

1.2.1 Aggregating data by Rider type and Bike type.

  • Aggregating data to see “Average minutes per ride” grouped by “bike type” and “rider type” after removing rides less than 2 minutes (As rides less than 2 minutes tend to have the same start and stop stations).
data_one_year_aggregate <- data_one_year %>% 
  select(ride_id, rideable_type, member_casual, started_at, ended_at,
         ride_length, everything()) %>%
  filter(ride_length >= 2) %>% 
  summarise("Number of Rides" = n(),
            "Ride Length" = sum(ride_length, na.rm = TRUE),
            "Max Ride Length" = round(max(ride_length), 2),
            "Avg Ride Length in Minutes" = round(mean(ride_length), 2),
            .by = c(member_casual, rideable_type)) %>% 
  arrange(desc("Avg Ride Length in Minutes")) %>% 
  gt() %>% tab_header(title = "Average length of Rides") %>% 
  cols_label(member_casual = "Rider type",
             rideable_type = "Bike type")

data_one_year_aggregate
Table 1:

Average minutes per ride
Average length of Rides
Rider type Bike type Number of Rides Ride Length Max Ride Length Avg Ride Length in Minutes
member classic_bike 1630991 21996488 1497.87 13.49
casual classic_bike 781530 19383358 1497.75 24.80
casual electric_bike 709649 11372659 479.98 16.03
member electric_bike 984688 10968684 480.00 11.14
casual docked_bike 136794 6899998 32035.45 50.44

We can clearly notice in Table 1 that member riders have more number of rides with both classic and electric bikes while the average ride length is higher with casual riders.

  • Calculating and visualizing Average ride length by “Rider type”.
average_ride_by_rideable_type <- data_one_year %>%
  rename("Rider type" = member_casual, "Bike type" = rideable_type) %>% 
  summarise(ride_length = sum(ride_length, na.rm = TRUE),
            ride_count = n(),
            avg_ride_length = ride_length/ride_count,
            .by = c(`Rider type`, `Bike type`)) %>% 
  ggplot(aes(`Rider type`, avg_ride_length)) + 
  geom_col(aes(fill = `Bike type`), position = "dodge") + 
  labs(x = "Bike type", y = "Avg Length of Ride(Minutes)",
       title  = "Average ride length by Bike type") +
  theme_minimal() +
  theme(plot.title = element_text(size = 18),
        legend.position = "bottom")

average_ride_by_rideable_type

Figure 1: Average Ride Length by Rider type and Member type

The above Figure 1 clearly shows that members average ride lengths between bike types doesn’t differ much for member riders but differs with casual riders upto 8 minutes.

Note

Further down in the analysis “docked_bike” type is dropped as no proper documentation is available in the course.

1.2.2 Analysing data by Time of the year and Ride Length

Ride Patterns Across the Weeks and Months of the Year

  • Calculating and visualizing ride patterns in a week for number of rides.
rideable_order <- c("classic_bike", "electric_bike", "docked_bike")

rides_on_days <- data_one_year %>%
  filter(rideable_type != "docked_bike") %>%
  mutate(month = month(started_at, label = TRUE,
                       abbr = FALSE)) %>% 
  mutate(rideable_type = factor(rideable_type,
                                levels = rideable_order)) %>%  ggplot(aes(wday(started_at, label = TRUE, abbr = FALSE))) + 
  geom_bar(aes(fill = member_casual), position = "dodge") +
  facet_wrap(~month, nrow = 3) + 
  labs(x = "Day of the Week", y = "Number of rides",
       title = "Riding pattrens on Weekdays of each Month",
       subtitle = "From July-2022 to June-2023",
       fill = "Type of Rider") +
  theme_light() +
  theme(legend.position = "top",
        axis.text.x = element_text(angle = 45, hjust = 1),
        plot.title = element_text(size = 18))

rides_on_days

Figure 2: Riding pattrens in Weekdays of each Month

The above Figure 2 clearly shows how the number of rides change due to seasons. In winters the number of rides decrease very drastically may be because of temperature and snow. In Summers the number of rides are at its peak.

The number of rides driven by member riders are increases through the week especially in working week days but for casual riders the rides increase in the weekends. The Figure 2 shows number of rides on Saturdays and Sundays by casual members overtake membership riders in the months of July and August.

Comparing variation in ride lengths of average and total ride lengths by bike type.

Aggregating data for the visualization.

rides_on_days <- data_one_year %>%
  mutate(day = wday(started_at, label = TRUE, abbr = FALSE),
         month = month(started_at, label = TRUE, abbr = FALSE)) %>% 
  summarise(ride_count = n(),
            sum_ride_length = sum(ride_length, na.rm = TRUE),
            avg_ride_length = mean(ride_length, na.rm = TRUE),
            .by = c(month, day, member_casual))

rides_on_days 
# A tibble: 168 × 6
   month day       member_casual ride_count sum_ride_length avg_ride_length
   <ord> <ord>     <chr>              <int>           <dbl>           <dbl>
 1 July  Wednesday member             45928         603429.            13.1
 2 July  Sunday    member             44328         685475.            15.5
 3 July  Wednesday casual             31631         703360.            22.2
 4 July  Saturday  member             51852         816009.            15.7
 5 July  Thursday  casual             34895         759013.            21.8
 6 July  Friday    member             46398         614466.            13.2
 7 July  Friday    casual             41198         959168.            23.3
 8 July  Monday    casual             32960         917333.            27.8
 9 July  Monday    member             38746         530041.            13.7
10 July  Sunday    casual             59528        1714086.            28.8
# ℹ 158 more rows

Let’s visualize the aggregated data

rides_on_days_len <- rides_on_days %>%
  ggplot(aes(day, sum_ride_length))+
  geom_col(aes(fill = member_casual), position = "dodge")+
  facet_wrap(~month, ncol = 3)+
  labs(x = "Day of the Week", y = "Total Length of Rides (Minutes)",
       title = "Total Minutes driven by Riders",
       fill = "Type of Rider") +
  theme(legend.position = "top",
        axis.text.x = element_text(angle = 45, hjust = 1),
        plot.title = element_text(size = 18))

rides_on_days_len

Figure 3: Total Ride lengths through out the year by member types.
rides_on_days_len_avg <- rides_on_days %>%
  ggplot(aes(day, avg_ride_length))+
  geom_col(aes(fill = member_casual), position = "dodge")+
  facet_wrap(~month, ncol = 3) +
  labs(x = "Day of the Week", y = "Average Length of Rides (Minutes)",
       title = "Average Minutes driven by Riders",
       fill = "Type of Rider") +
  theme(legend.position = "top",
        axis.text.x = element_text(angle = 45, hjust = 1),
        plot.title = element_text(size = 18))

rides_on_days_len_avg

Figure 4: Average Ride lengths through out year by member types.

The ride length is varying across months and seasons just as number of rides but average ride length is not fluctuating that much across the year.

Rides through out the day

Let’s look at when the rides are starting to know at what time of day the rides peak and are at the lowest.

rides_on_time_of_day <- data_one_year %>%
  mutate(time_of_day = format(as.POSIXct(ended_at), "%H"),
         wk_day = wday(started_at, label = TRUE, abbr = FALSE)) %>% 
  summarise(ride_id = n(),
            .by = c(time_of_day, member_casual))

rides_on_time_of_day %>%
  ggplot(aes(time_of_day, ride_id, fill = ride_id )) +
  geom_col() +
  labs(x = "Hour of the day", y = "Number of Rides",
       fill = "Max Rides") +
  facet_wrap(~member_casual, ncol = 1) +
  scale_y_continuous(
  labels = scales::number_format(scale = 1e-3, suffix = "K")) +
  theme_minimal() +
  theme(legend.position = "none")

  • Most of the rides start at 5:00 PM in the evening showing most of the rides begin after office hours for both caual and member riders but members peak twice at 8:00 AM and 5:00 PM but the casual riders peak only once at 5:00 PM.

1.2.3 Analysing of Stations and Routes.

  • Removing “NA” and blanks from the stations columns.
data_one_year <- data_one_year_raw %>%
  mutate(ride_length = difftime(ended_at, started_at,
                                units = "min")) %>%
  mutate(ride_length = as.numeric(ride_length)) %>% 
  mutate(ride_length = if_else(ride_length < 0, 0, ride_length)) %>% 
  filter(ride_length >= 2) %>% 
  drop_na(start_station_name, end_station_name ) %>% 
  filter(start_station_name != "" & end_station_name != "",
         started_at != ended_at)

glimpse(data_one_year)
Rows: 4,243,662
Columns: 14
$ ride_id            <chr> "954144C2F67B1932", "57765852588AD6E0", "B5B6BE4431…
$ rideable_type      <chr> "classic_bike", "classic_bike", "classic_bike", "cl…
$ started_at         <dttm> 2022-07-05 08:12:47, 2022-07-03 13:58:49, 2022-07-…
$ ended_at           <dttm> 2022-07-05 08:24:32, 2022-07-03 14:06:32, 2022-07-…
$ start_station_name <chr> "Ashland Ave & Blackhawk St", "Buckingham Fountain …
$ start_station_id   <chr> "13224", "15541", "15541", "TA1307000117", "15535",…
$ end_station_name   <chr> "Kingsbury St & Kinzie St", "Michigan Ave & 8th St"…
$ end_station_id     <chr> "KA1503000043", "623", "TA1307000164", "TA130700005…
$ start_lat          <dbl> 41.90707, 41.86962, 41.86962, 41.89147, 41.88461, 4…
$ start_lng          <dbl> -87.66725, -87.62398, -87.62398, -87.62676, -87.644…
$ end_lat            <dbl> 41.88918, 41.87277, 41.79526, 41.93625, 41.86712, 4…
$ end_lng            <dbl> -87.63851, -87.62398, -87.59647, -87.65266, -87.641…
$ member_casual      <chr> "member", "casual", "casual", "member", "member", "…
$ ride_length        <dbl> 11.750000, 7.716667, 58.483333, 26.300000, 8.716667…
  • Making a new column to identify travelled routes.
data_one_year <- data_one_year %>% 
  mutate(stations_travelled = paste(start_station_name, 
                                     "-", end_station_name))

glimpse(data_one_year)
Rows: 4,243,662
Columns: 15
$ ride_id            <chr> "954144C2F67B1932", "57765852588AD6E0", "B5B6BE4431…
$ rideable_type      <chr> "classic_bike", "classic_bike", "classic_bike", "cl…
$ started_at         <dttm> 2022-07-05 08:12:47, 2022-07-03 13:58:49, 2022-07-…
$ ended_at           <dttm> 2022-07-05 08:24:32, 2022-07-03 14:06:32, 2022-07-…
$ start_station_name <chr> "Ashland Ave & Blackhawk St", "Buckingham Fountain …
$ start_station_id   <chr> "13224", "15541", "15541", "TA1307000117", "15535",…
$ end_station_name   <chr> "Kingsbury St & Kinzie St", "Michigan Ave & 8th St"…
$ end_station_id     <chr> "KA1503000043", "623", "TA1307000164", "TA130700005…
$ start_lat          <dbl> 41.90707, 41.86962, 41.86962, 41.89147, 41.88461, 4…
$ start_lng          <dbl> -87.66725, -87.62398, -87.62398, -87.62676, -87.644…
$ end_lat            <dbl> 41.88918, 41.87277, 41.79526, 41.93625, 41.86712, 4…
$ end_lng            <dbl> -87.63851, -87.62398, -87.59647, -87.65266, -87.641…
$ member_casual      <chr> "member", "casual", "casual", "member", "member", "…
$ ride_length        <dbl> 11.750000, 7.716667, 58.483333, 26.300000, 8.716667…
$ stations_travelled <chr> "Ashland Ave & Blackhawk St - Kingsbury St & Kinzie…
  • Finding which route is most traveled by casual riders.
most_travelled_routes_casual <- data_one_year %>%
  filter(member_casual == "casual",
         ride_length >= 2) %>% 
  summarise(ride_count = n(),
            avg_ride_length = round(mean(ride_length), 2),
            .by = c(stations_travelled)) %>%
  arrange(desc(ride_count))

head(most_travelled_routes_casual)
# A tibble: 6 × 3
  stations_travelled                                  ride_count avg_ride_length
  <chr>                                                    <int>           <dbl>
1 Streeter Dr & Grand Ave - Streeter Dr & Grand Ave         8259            46.3
2 DuSable Lake Shore Dr & Monroe St - DuSable Lake S…       5726            38.3
3 DuSable Lake Shore Dr & Monroe St - Streeter Dr & …       4840            27.1
4 Michigan Ave & Oak St - Michigan Ave & Oak St             3754            50.9
5 Millennium Park - Millennium Park                         3188            45.4
6 Streeter Dr & Grand Ave - DuSable Lake Shore Dr & …       2663            27.8
NROW(most_travelled_routes_casual)
[1] 130373

Streeter Dr & Grand Ave - Streeter Dr & Grand Ave stands to be the most popular station with 9698 rides by casual riders.

most_travelled_routes_member <- data_one_year  %>%
  filter(member_casual == "member") %>% 
  summarise(ride_count = n(),
            total_ride_length = sum(ride_length),
            ride_length = round(mean(ride_length), 2),
            .by = stations_travelled) %>% arrange(desc(ride_count))

head(most_travelled_routes_member)
# A tibble: 6 × 4
  stations_travelled                    ride_count total_ride_length ride_length
  <chr>                                      <int>             <dbl>       <dbl>
1 Ellis Ave & 60th St - University Ave…       6149            25928.        4.22
2 University Ave & 57th St - Ellis Ave…       5771            26605.        4.61
3 Ellis Ave & 60th St - Ellis Ave & 55…       5676            28427.        5.01
4 Ellis Ave & 55th St - Ellis Ave & 60…       5347            27187.        5.08
5 State St & 33rd St - Calumet Ave & 3…       4037            17795.        4.41
6 Calumet Ave & 33rd St - State St & 3…       3836            15538.        4.05
NROW(most_travelled_routes_member)
[1] 144802

Ellis Ave & 60th St - University Ave & 57th St stands as the most traveled route by member riders with 6153 rides per anum.

  • Finding which station has most ride starting points and which station has most ending points.
most_starting_points <- data_one_year %>% 
  summarise(ride_count = n(),
            .by = start_station_name) %>%
  select(start_station_name, ride_count) %>%
  slice_max(ride_count, n = 10)

most_starting_points
# A tibble: 10 × 2
   start_station_name                 ride_count
   <chr>                                   <int>
 1 Streeter Dr & Grand Ave                 63899
 2 DuSable Lake Shore Dr & Monroe St       36757
 3 Michigan Ave & Oak St                   35050
 4 DuSable Lake Shore Dr & North Blvd      34167
 5 Wells St & Concord Ln                   32175
 6 Clark St & Elm St                       31832
 7 Kingsbury St & Kinzie St                30820
 8 Millennium Park                         29894
 9 Theater on the Lake                     28864
10 Wells St & Elm St                       27152
most_starting_points$ride_count %>% sum()
[1] 350610
most_ending_points <- data_one_year %>% 
  summarise(ride_count = n(),
            .by = end_station_name) %>%
  select(end_station_name, ride_count)  %>% 
  slice_max(ride_count, n = 10)

most_ending_points
# A tibble: 10 × 2
   end_station_name                   ride_count
   <chr>                                   <int>
 1 Streeter Dr & Grand Ave                 65542
 2 DuSable Lake Shore Dr & North Blvd      37093
 3 Michigan Ave & Oak St                   35977
 4 DuSable Lake Shore Dr & Monroe St       35629
 5 Wells St & Concord Ln                   32879
 6 Clark St & Elm St                       31394
 7 Millennium Park                         31023
 8 Kingsbury St & Kinzie St                29805
 9 Theater on the Lake                     29482
10 Wells St & Elm St                       27360
most_ending_points$ride_count %>% sum()
[1] 356184

Streeter Dr & Grand Ave found to be the most popular station as most rides start and end at that station.

1.2.4 Looking at Filtered data

Just because we filtered the data with NA’s that does not mean that the data is not helpful, it just means that it does not our fulfill specific need when calculating or manipulating data.

Let’s look at NA’s in the data once again.

na_in_cols <- data_one_year_raw %>% map( ~sum(is.na(.))) %>% unlist()

na_in_cols
           ride_id      rideable_type         started_at           ended_at 
                 0                  0                  0                  0 
start_station_name   start_station_id   end_station_name     end_station_id 
            857860             857992             915655             915796 
         start_lat          start_lng            end_lat            end_lng 
                 0                  0               5795               5795 
     member_casual 
                 0
  • We can see that the start_station_name and end_station_name have majority of NA’s it means that rides are starting and ending where stations are not there.
prop_na <- na_in_cols["start_station_name"]/nrow(data_one_year_raw)

prop_na
start_station_name 
          0.148433
  • 14.84% of data in start_station_name is missing and good thing is that none of the start_lng and start_lat have any NA’s and we can use this for find the most traveled routes.
data_na_one_year <- data_one_year_raw %>% 
  filter(is.na(start_station_name) | start_station_name == "") %>% 
  drop_na(end_lat, end_lng)
  
glimpse(data_na_one_year)
Rows: 857,860
Columns: 13
$ ride_id            <chr> "DCB3D2C9B63999EC", "D1ACA8280DA02AE3", "EF98673429…
$ rideable_type      <chr> "electric_bike", "electric_bike", "electric_bike", …
$ started_at         <dttm> 2022-07-04 15:04:26, 2022-07-12 14:43:51, 2022-07-…
$ ended_at           <dttm> 2022-07-04 15:32:38, 2022-07-12 14:49:28, 2022-07-…
$ start_station_name <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ start_station_id   <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ end_station_name   <chr> "Ashland Ave & Blackhawk St", "Cornell Ave & Hyde P…
$ end_station_id     <chr> "13224", "KA1503000007", "KA1503000007", "847", "48…
$ start_lat          <dbl> 41.95, 41.80, 41.80, 41.74, 42.02, 41.95, 41.95, 41…
$ start_lng          <dbl> -87.64, -87.59, -87.59, -87.55, -87.69, -87.67, -87…
$ end_lat            <dbl> 41.90707, 41.80241, 41.80241, 41.73000, 42.01000, 4…
$ end_lng            <dbl> -87.66725, -87.58692, -87.58692, -87.55000, -87.690…
$ member_casual      <chr> "member", "member", "member", "member", "member", "…
  • Now let’s make new columns start_point with start_lng and start_lat and end_point with end_lat and end_lng.
data_na_one_year <- data_na_one_year %>%
    mutate(start_point = paste(start_lat, start_lng),
           end_point = paste(end_lat, end_lng))

glimpse(data_na_one_year)
Rows: 857,860
Columns: 15
$ ride_id            <chr> "DCB3D2C9B63999EC", "D1ACA8280DA02AE3", "EF98673429…
$ rideable_type      <chr> "electric_bike", "electric_bike", "electric_bike", …
$ started_at         <dttm> 2022-07-04 15:04:26, 2022-07-12 14:43:51, 2022-07-…
$ ended_at           <dttm> 2022-07-04 15:32:38, 2022-07-12 14:49:28, 2022-07-…
$ start_station_name <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ start_station_id   <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,…
$ end_station_name   <chr> "Ashland Ave & Blackhawk St", "Cornell Ave & Hyde P…
$ end_station_id     <chr> "13224", "KA1503000007", "KA1503000007", "847", "48…
$ start_lat          <dbl> 41.95, 41.80, 41.80, 41.74, 42.02, 41.95, 41.95, 41…
$ start_lng          <dbl> -87.64, -87.59, -87.59, -87.55, -87.69, -87.67, -87…
$ end_lat            <dbl> 41.90707, 41.80241, 41.80241, 41.73000, 42.01000, 4…
$ end_lng            <dbl> -87.66725, -87.58692, -87.58692, -87.55000, -87.690…
$ member_casual      <chr> "member", "member", "member", "member", "member", "…
$ start_point        <chr> "41.95 -87.64", "41.8 -87.59", "41.8 -87.59", "41.7…
$ end_point          <chr> "41.907066 -87.667252", "41.802406 -87.586924", "41…
  • Aggregating data to check for the most traveled routes without a start_station name.

First join start_point and end_point to make route_travelled then count the rides by routes_travelled to see the most traveled path.

most_travelled_na_routes <- data_na_one_year %>%
  filter(start_point != end_point) %>% 
  mutate(route_travelled = paste(start_point, ",", end_point)) %>% 
  summarise(ride_count = n(),
            .by = route_travelled) %>%
  slice_max(ride_count, n=10)

most_travelled_na_routes
# A tibble: 10 × 2
   route_travelled                             ride_count
   <chr>                                            <int>
 1 41.79 -87.6 , 41.8 -87.59                         1459
 2 41.79 -87.59 , 41.79 -87.6                        1354
 3 41.8 -87.59 , 41.79 -87.6                         1335
 4 41.79 -87.6 , 41.79 -87.59                        1320
 5 41.8 -87.6 , 41.79 -87.6                          1099
 6 41.79 -87.6 , 41.78509714636 -87.6010727606       1058
 7 41.79 -87.6 , 41.8 -87.6                           999
 8 41.79 -87.6 , 41.799568 -87.594747                 917
 9 41.79 -87.6 , 41.78 -87.6                          697
10 41.89 -87.63 , 41.9 -87.63                         690
sum(most_travelled_na_routes["ride_count"])
[1] 10928
  • 10928 rides are not small when compared to most traveled routes, but 10928 rides in 5 million rides is not that high.

1.3 Conclusion

The analysis of the cyclist trip data reveals several key insights:

1.3.1 Rider Patterns

  • Member riders tend to use bikes more frequently than casual riders, especially during weekdays.
  • Casual riders show a preference for weekends, with a significant increase in rides during Saturdays and Sundays.

1.3.2 Ride Length

  • The average ride length is generally longer for casual riders compared to member riders.
  • The longest average ride lengths occur on weekends, particularly for casual riders.

1.3.4 Temporal Patterns

  • Ride patterns vary by time of day, with peak usage in the morning and evening hours.
  • The analysis highlights the importance of understanding temporal patterns to optimize bike availability and station placements.

1.4 Recommendations for Improving the Ride-Share System

1.4.1 1. Rider Patterns

  • Launch membership campaigns to convert casual weekend riders into members through discounts or referral incentives.
  • Introduce weekend promotions such as family/group ride offers to capture higher weekend demand.

1.4.2 2. Ride Length

  • Implement flexible pricing tiers for longer rides, especially for casual riders, to encourage higher usage.
  • Form tourist and leisure partnerships (e.g., with parks, attractions, waterfronts) to promote longer leisure rides.

1.4.4 4. Temporal Patterns

  • Ensure high bike availability during morning and evening commute peaks to support consistent member usage.
  • Use dynamic bike rebalancing strategies to redistribute bikes based on time-of-day demand (e.g., office hubs, transit points).
  • Introduce flexible commuter plans (weekly/monthly passes) tailored for regular peak-hour riders.

1.4.5 Overall Strategy

  • Focus on converting casual riders into members with targeted offers.
  • Leverage seasonal and temporal insights to optimize bike availability and station placement.
  • Apply differentiated pricing and promotions to encourage longer and more frequent rides.