Analyzing the bike sharing data from Kaggle Datasets
Top routes for bike sharing consists of distances that ranges from 0.716 km to 2.798 km.
Ellis Ave 60th stto55th stroute ranks the most number of rides (5k), with the average of 1.022 km.Shedd aquariumtostreeter Dr. and grand averoute has the highest distance, with an average of 3050 rides.
- Ridership for weekdays has two peaks:
- 8 AM between 1112 to 1494 rides
- 5 PM between 2523 to 2969 rides
- Distance travelled during weekdays are right-skewed.
- Ridership for weekend are partiall normal with a wider peak. For both Saturday and Sunday bikers has longer hour of activities which started from 10 AM to 7 PM.
- Additionally, during weekends, travel distance can reach up to 2.4 km in 4 AM to 2.3 km in 5 PM, that peaks at 2pm with 2.5 km.
- The three bike types follow almost the same distance distribution, with the exception of electric bikes because of its outlier.
- Electric bikes scan be used to transport up to 100 km distance.
- Negative velocity for electric and classic suggests that they can be used from origin to destination then vice versa. We can infer, in a network perspective that transportation in this two type of bikes are bidirectional in nature.
- On the other hand, docked bikes are unidirectional.
| Null Hypthesis (HO) | Alternative Hypothesis (HA) | p-value | Conlusion |
|---|---|---|---|
| The mean travel time is the same for electric bikes, classic and docked bikes |
The mean travel time is different for electric bikes, classic and docked bikes | 1.141772e-10 | Travel time is dependent to bicycle type |
| There is no significant difference in travel distance between members and non-members |
There is a significant difference in travel distance between members and non-members |
0.748 | There is no significant difference between the distance travelled by member or casual |
| Travel distance is independent of the type of bike | Travel distance depends on the type of bike | 0.076429 | distance is independent from bicycle type |
| The average travel time is the same between peak and non-peak hours |
The average travel time is the same between peak and non-peak hours |
0.106 | average travel time during weekdays is the same between peak and non-peak hours |
| The average travel time is the same between peak and non-peak hours |
The average travel time is the same between peak and non-peak hours |
0.074 | average travel time during weekends is the same between peak and non-peak hours |
| Bike membership status does not influence the choice of bike type |
Bike membership status influences the choice of bike type | 4.929534e-14 | There is an association between membership type and bicycle type |
| There is no interaction effect between bike type and membership status on travel velocity |
There is an interaction effect between bike type and membership status on travel velocity |
3.171383e-23 | Travel velocity is affected by membership and bicycle type |