The purpose of the analyses is to create a summary DataFrame of the ride-sharing data by city type. Next, the summary data is going to be used for creating a multiple-line graph that illustrates the total weekly fares for each city type by using Pandas and Matplotlib tools. Finally, the graph sheds light on the data to explain how the data differs by city type and how those differences can be used by decision-makers at PyBer.
According to the data, the following analyses were created,
1. PyBer Ride-Sharing Data (2019)
2. Ride Count Data (2019)
3. Ride Fare Data (2019)
4. Driver Count Data (2019)
5. % of Total Fares by City Type
6. % of Total Rides by City Type
7. % of Total Drivers by City Type
8. Total Fare by City Type
A bubble chart below shows PyBer ride-sharing data by city types. Urban areas have the highest total number of rides than suburban and rural areas. On the contrary, the lowest average fare was seen in the rural regions.
The box-and-whisker plot shows the number of rides by city types. According to the graph, there is at least one outlier in urban ride count, which is close to 40. Further, the total number of trips in rural areas is nearly 4- and 3.5 times smaller per area than the urban and suburban areas, respectively.
The box-and-whisker plot illustrates the ride fare data by city types. According to the graph, there is no outliers. Therefore, the average fare for rides in the rural cities is about $37, which means around $11 and $5 more per ride than the urban and suburban cities, respectively.
The box-and-whisker plot indicates the number of drivers by city types. The average number of drivers in rural cities is nine to four times less per city than in urban and suburban cities, respectively.
The pie chart shows the percentage of total fares by city type. The total fares in urban cities is two to nine times greater per city than in suburban and rural cities.
The pie chart illustrates the percentage of total ride by city type. The total rides in urban cities is 2.5 and 13 times higher per city than in suburban and rural cities.
The final pie graph presents the percentage of total drivers by city type. 80.9 % of drivers are in urban cities, which is the highest number of driver ratio per city.
Finally, the multi-line chart shows total fare by city types. It indicated a correlation about density which defines the relationship between fare/ride and population. As we know that according to the census, in general, a big portion of the population lives in urban areas. This idea illustrates more rides and more fares in urban areas. As a result, The PyBer gets most of its revenue from urban areas, which have high population density.
Based on the analyses, in my opinion, first, they should increase the number of drivers in the rural cities, which should cover the driver demand in the specific area. Second, the analyses show that the average fare per ride and driver in rural cities is much greater than in suburban and urban cities. To balance the average fare and rural area drivers' demand, which needs more analyzing at the local scale, and which may be adding the population variable into the neighbourhood scale and doing more analyzes.