Investigation Overview¶
I started out just to finish my nanodegree program but as I began to explore, the dataset started to reveal intresting insights to me hence the listed points were noted:
- The duration_sec is not normally distributed but skewed. Then I create new feature travel_minutes to enable me observe it more closely which shows that the majority of the members do not use bike share for their trips, and most were around 25 to 40 years old. While Most rides were quick and short, that lasted between 5 to 20 minutes, some riders travel time is close to 23hrs which is not normal.
- There is significant difference in the avagrage time travel by customer(13mins) and subscribers(8mins). where is seems subscribers tend to know where they going hereby reducing travel time.
- the riding patterns between young riders and older riders can both be similar and at the same time be different, depending on the feature selected as microscope.
I hope this piece of work be of assistance to as many aspiring data analyst as possible. :airplane #
Dataset Overview¶
Bay Wheels (previously known as Ford GoBike) is a regional public bike sharing system in the San Francisco Bay Area, California. Bay Wheels is the first regional and large-scale bicycle sharing system deployed in California and on the West Coast of the United States with nearly 500,000 rides since the launch in 2017 and had about 10,000 annual subscribers as of January 2018. The dataset used for this exploratory analysis consists of This dataset includes information about individual rides made in a bike-sharing system covering the Francisco Bay area of travels for the month February year 2019 for February 2019 alone in CSV format covering the greater San Francisco Bay area, also available here data for other cities.
This dataset initially consists of a total of 183412 rows and 16 columns where six of th 16 columns are having null values you can check the cell above for the amount of null entries for each feature. Then after I had wrangled the data it has reduced to 174952 rows and the columns increased to 21 columns. This dataset is for rides in the month of february 2019
(Initial Visualizations)¶
Usually most rides take about 20mins ride from a station to another and while the travel time in hours shows that usually the time during rides are
within one hour from boarding station to alighting station.
The travel_minutes shows that the majority of the members did not use bike share for all of their trips, and most were around 25 to 40 years old. Most rides were quick and short, lasted between 5 to 20 minutes, that some riders travel time is close to 23hrs which is not normal.
Irregular travel time upto 23hrs¶
| Unnamed: 0 | duration_sec | start_time | end_time | start_station_id | start_station_name | start_station_latitude | start_station_longitude | end_station_id | end_station_name | ... | bike_share_for_all_trip | year | month | days | boarding_hour | period_of_day | day_of_week | age | travelTime_minutes | travelTime_hours | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 4987 | 5203 | 83195 | 2019-02-27 14:47:23.181 | 2019-02-28 13:53:58.433 | 243.0 | Bancroft Way at College Ave | 37.869360 | -122.254337 | 248.0 | Telegraph Ave at Ashby Ave | ... | Yes | 2019 | 2 | 27 | 14 | afternoon | Tuesday | 57 | 1386 | 23 |
| 81604 | 85465 | 84548 | 2019-02-16 15:48:25.029 | 2019-02-17 15:17:33.080 | 3.0 | Powell St BART Station (Market St at 4th St) | 37.786375 | -122.404904 | 368.0 | Myrtle St at Polk St | ... | No | 2019 | 2 | 16 | 15 | afternoon | Friday | 38 | 1409 | 23 |
| 107291 | 112435 | 83407 | 2019-02-11 16:25:33.069 | 2019-02-12 15:35:40.956 | 77.0 | 11th St at Natoma St | 37.773507 | -122.416040 | 344.0 | 16th St Depot | ... | No | 2019 | 2 | 11 | 16 | afternoon | Sunday | 31 | 1390 | 23 |
| 122163 | 127999 | 83519 | 2019-02-09 15:16:17.537 | 2019-02-10 14:28:17.270 | 72.0 | Page St at Scott St | 37.772406 | -122.435650 | 43.0 | San Francisco Public Library (Grove St at Hyde... | ... | No | 2019 | 2 | 9 | 15 | afternoon | Friday | 29 | 1391 | 23 |
4 rows × 26 columns
(Visualizations For More Insights)¶
Female riders tends to slightly spend more time riding than other genders.
There is significant difference in the avagrage time travel by customer(13mins) and subscribers(8mins). where is seems subscribers tend to know where they going hereby reducing travel time.
The busiest hours are pretty much consistent throughout the month with rush hours occuring around 8AM and 5PM.
(Visualizations With Little More Nuances)¶
Wednesday is the day the ride service is used the most whilst during the weekend there is a constant reduction of usage expectially Fridays and Saturdays.
That dataset also shows that even though all genders incresed their travel time during weekends, male riders travel time reduces by most significantly on saturday.
older customers tend to spend far less time riding during weekends on like youth customers but older subcribers spend similar duration of time younger subcribers spend during weekend.
looking at people where ages >60yrs other gender spend less time riding from thursday-friday than other genders where male users i.e (sat-sun) looks exactly like that of the younger male users both female users spend more time riding from wednesay- Friday.
older Female riders tend to spend more time riding than male riders.
male users tend to use the platform more than other genders and throughout the day on like other genders that usage of the platform is well pronounced from 5am.
