TfL Bikes Hired

With the society becoming more and more aware of what is affecting the environment, many individuals are shifting to bike usage as compared to cars. With this project, we want to analyse what could be the factors that affect increase or decrease in bike rentals in London.

Let’s have a look at the data

skimr::skim(bike0)

(#tab:skim_data)Data summary
Name	bike0
Number of rows	4416
Number of columns	2
_______________________
Column type frequency:
numeric	1
POSIXct	1
________________________
Group variables	None

Variable type: numeric

skim_variable	n_missing	complete_rate	mean	sd	p0	p25	p50	p75	p100	hist
Number of Bicycle Hires	0	1	26844	9900	2764	19698	26607	34206	73094	▃▇▅▁▁

Variable type: POSIXct

skim_variable	n_missing	complete_rate	min	max	median	n_unique
Day	0	1	2010-07-30	2022-08-31	2016-08-14 12:00:00	4416

First, we need to clean the data to make it easier to perform data visualisation on it.

# change dates to get year, month, and week
bike <- bike0 %>% 
  clean_names() %>% 
  rename (bikes_hired = number_of_bicycle_hires) %>% 
  mutate (year = year(day),
          month = lubridate::month(day, label = TRUE),
          week = isoweek(day))

How many bikes were hired per month and year since 2015?

We plot a density plot to find out the distribution of the bikes per month since 2015.

Next, we try to find the difference in expected and actual bikes hired in London between 2017-2022. Here, we have calculated expected based on the monthly averages between 2016-2019.

One important thing to note in this graph is that since, 2016-2019 average values are included in calculating the expected bikes hired, the line for the actual bikes hired does not vary too much from the expected. But, for 2020 to 2022, we can clearly observe a lot more variations.

Now, we do the same thing but keeping in mind the weekly averages of bikes hired.

We decided to use mean, because the data showed seasonal trends and there were not many clear outliers, which would affect the data.

Moreover, from the data we observed that the significant highs or lows were attributed to certain events that took place during the time period. Furthermore, the COVID-19 pandemic was one of the reasons that we saw decrease in hires as the country was in lock-down. But, once the country opened up we observed that people wanted to avoid the crowded public transport due to fear of infection and hence, resorted to biking.