Systems and methods for generating mobility insurance products using ride-sharing telematics data

1. A personalized insurance (“PI”) computing device for developing a model to determine a ratings-based insurance product for a driver of a transportation network company (TNC) operating a vehicle for the TNC, the PI computing device in communication with a TNC computing device associated with the TNC, a plurality of sensors via a vehicle computing device associated with the vehicle, and a financial institution that manages a health savings account for the driver, the PI computing device having at least one processor in communication with at least one memory, the at least one processor configured to:

retrieve ratings for the driver and the vehicle operated by the driver;

receive telematics data from the vehicle computing device, the telematics data associated with operation of the vehicle and collected by the plurality of sensors;

generate, using one or more machine learning programs, a risk model for the driver based upon the ratings and the telematics data;

receive additional ratings for the driver and the vehicle, additional telematics data associated with current operation of the vehicle, and environment data indicative of current environmental conditions of an area where the vehicle is being operated;

receive, in real-time from the TNC computing device, supply and demand data indicative of current supply and demand for the TNC;

train, using the one or more machine learning programs, the risk model by applying the additional telematics data, the environment data, the supply and demand data, and the additional ratings to the risk model;

determine, in real-time, a personalized optimal insurance product for the driver by executing the trained risk model, wherein the personalized optimal insurance product is personalized for the driver and vehicle;

transmit, to a first user computing device associated with the driver, the personalized optimal insurance product to the driver;

receive, from the first user computing device, a message indicating a purchase of the personalized optimal insurance product; and

transmit a message to the financial institution to withdraw an amount of funds from the health savings account for the purchase of the personalized optimal insurance product.

2. The PI computing device of claim 1 , wherein the at least one processor is further configured to:

receive the ratings from a second user computing device associated with a passenger, wherein the ratings are associated with a ride provided to the passenger, and wherein the ratings include at least one numerical score associated with the ride provided by the driver; and

store the ratings in the memory.

3. The PI computing device of claim 1 , wherein the ratings include a description of a ride provided by the driver, wherein the description is inputted by a passenger of the ride, and wherein the at least one processor is further configured to:

parse the description using natural language processing; and

analyze the parsed description to determine a numerical score corresponding to a rating for the ride provided by the driver.

4. The PI computing device of claim 1 , wherein the ratings and the additional ratings include performance metrics for at least one ride provided to a passenger, the performance metrics retrieved from measurement devices on the vehicle, the performance metrics including the telematics data, the telematics data including at least one of acceleration data, braking data, and cornering data, and wherein the ratings and the additional ratings further include ratings associated with the vehicle, ratings associated with the driver, and location data associated with trips performed by the driver while operating the vehicle.

5. The PI computing device of claim 1 , wherein generating the risk model includes overlaying the retrieved ratings with data associated with a geographic region associated with a route for a ride provided to a passenger.

6. The PI computing device of claim 1 , wherein generating the risk model includes incorporating an optimal pricing model, and wherein the at least one processor is further configured to:

retrieve, from the TNC, data indicating a level of demand for transportation services; and

generate the optimal pricing model based upon the retrieved demand data.

7. The PI computing device of claim 1 , wherein the environment data is one of (a) detected by the plurality of sensors and received from the vehicle computing device, and (b) received from at least one environment data source.

8. The PI computing device of claim 1 , wherein the environment data includes at least one weather related data or traffic conditions.

9. A computer-implemented method for developing a model to determine a ratings-based insurance product for a driver of a transportation network company (TNC) operating a vehicle for the TNC using a personalized insurance (“PI”) computing device, the PI computing device in communication with TNC computing device associated with the TNC, a plurality of sensors via a vehicle computing device associated with the vehicle, and a financial institution that manages a health savings account for the driver, the PI computing device having at least one processor in communication with at least one memory, the method comprising:

retrieving ratings for the driver and the vehicle operated by the driver;

receiving telematics data from the vehicle computing device, the telematics data associated with operation of the vehicle and collected by the plurality of sensors;

generating, using one or more machine learning programs, a risk model for the driver based upon the ratings and the telematics data;

receiving additional ratings for the driver and the vehicle, additional telematics data associated with current operation of the vehicle, and environment data indicative of current environmental conditions of an area where the vehicle is being operated;

receiving, in real-time from the TNC computing device, supply and demand data indicative of current supply and demand for the TNC;

training, using the one or more machine learning programs, the risk model by applying the additional telematics data, the environment data, the supply and demand data, and the additional ratings to the risk model;

determining, in real-time, a personalized optimal insurance product for the driver by executing the trained risk model, wherein the personalized optimal insurance product is personalized for the driver and vehicle;

transmitting, to a first user computing device associated with the driver, the personalized optimal insurance product to the driver;

receiving, from the first user computing device, a message indicating a purchase of the personalized optimal insurance product; and

transmitting a message to the financial institution to withdraw an amount of funds from the health savings account for the purchase of the personalized optimal insurance product.

10. The computer-implemented method of claim 9 , the method further comprising:

receiving the ratings from a second user computing device associated with a passenger, wherein the ratings are associated with a ride provided to the passenger, and wherein the ratings include at least one numerical score associated with the ride provided by the driver; and

storing the ratings in the memory.

11. The computer-implemented method of claim 9 , wherein the ratings include a description of a ride provided by the driver, the description inputted by a passenger of the ride provided, the method further comprising:

parsing the description using natural language processing; and

analyzing the parsed description to determine a numerical score corresponding to a rating for the ride provided by the driver.

12. The computer-implemented method of claim 9 , wherein the ratings and the additional ratings include performance metrics for at least one ride provided to a passenger, the performance metrics retrieved from measurement devices on the vehicle, the performance metrics including the telematics data, the telematics data including at least one of acceleration data, braking data, and cornering data, and wherein the ratings and the additional ratings further include ratings associated with the vehicle, ratings associated with the driver, and location data associated with trips performed by the driver while operating the vehicle.

13. The computer-implemented method of claim 9 , the method further comprising generating the risk model further based on overlaying the retrieved ratings with data associated with a geographic region associated with a route for a ride provided to a passenger.

14. The computer-implemented method of claim 9 , the method further comprising:

retrieving, from the TNC, data indicating a level of demand for transportation services;

generating an optimal pricing model based upon the retrieved demand data; and

generating the risk model further based upon the generated optimal pricing model.

15. One or more non-transitory computer-readable storage media having computer-executable instructions embodied thereon, when executed by a personalized insurance (“PI”) computing device in communication with a transportation network company (TNC) computing device associated with a TNC, a plurality of sensors via a vehicle computing device associated with a vehicle, and a financial institution that manages a health savings account for a driver, the PI computing device having at least one processor in communication with at least one memory, the computer-executable instructions causing the at least one processor to:

retrieve ratings for the driver and the vehicle operated by the driver;

receive telematics data from the vehicle computing device, the telematics data associated with operation of the vehicle and collected by the plurality of sensors;

generate, using one or more machine learning programs, a risk model for the driver based upon the ratings and the telematics data;

receive additional ratings for the driver and the vehicle, additional telematics data associated with current operation of the vehicle, and environment data indicative of current environmental conditions of an area where the vehicle is being operated;

receive, in real-time from the TNC computing device, supply and demand data indicative of current supply and demand for the TNC;

train, using the one or more machine learning programs, the risk model by applying the additional telematics data, the environment data, the supply and demand data, and the additional ratings to the risk model;

determine, in real-time, a personalized optimal insurance product for the driver by executing the risk model, wherein the personalized optimal insurance product is personalized for the driver and vehicle;

transmit, to a first user computing device associated with the driver, the personalized optimal insurance product to the driver;

receive, from the first user computing device associated with the driver, a message indicating a purchase of the personalized optimal insurance product; and

transmit a message to the financial institution to withdraw an amount of funds from the health savings account for the purchase of the personalized optimal insurance product.

16. The non-transitory computer-readable storage media of claim 15 , wherein the computer-executable instructions further cause the at least one processor to:

receive the ratings from a second user computing device associated with a passenger, wherein the ratings are associated with a ride provided to the passenger, and wherein the ratings include at least one numerical score associated with the ride provided by the driver; and

store the ratings in the memory.

17. The non-transitory computer-readable storage media of claim 15 , wherein the ratings include a description of a ride provided by the driver, wherein the description is inputted by a passenger of the ride, and wherein the computer-executable instructions further cause the at least one processor to:

parse the description using natural language processing; and

analyze the parsed description to determine a numerical score corresponding to a rating for the ride provided by the driver.

18. The non-transitory computer-readable storage media of claim 15 , wherein the ratings and the additional ratings include performance metrics for at least one ride provided to a passenger, the performance metrics retrieved from measurement devices on a vehicle associated with the driver, the performance metrics including the telematics data, the telematics data including at least one of acceleration data, braking data, and cornering data, and wherein the ratings and the additional ratings further include ratings associated with the vehicle, ratings associated with the driver, and location data associated with trips performed by the driver while operating the vehicle.

19. The non-transitory computer-readable storage media of claim 15 , wherein generating the risk model includes overlaying the retrieved ratings with data associated with a geographic region associated with a route for a ride provided to a passenger.

20. The non-transitory computer-readable storage media of claim 15 , wherein the computer-executable instructions further cause the at least one processor to:

retrieve, from the TNC, data indicating a level of demand for transportation services;

generate an optimal pricing model based upon the retrieved demand data; and

generate the risk model further based upon the generated optimal pricing model.