Pre-selection of drivers in a passenger transport system

• US 10,417,589 B2
• Filed: 11/30/2016
• Issued: 09/17/2019
• Est. Priority Date: 11/01/2016
• Status: Active Grant

• Current Assignee: Uber Technologies, Inc.
• Original Assignee: Uber Technologies, Inc.
• Inventors: Zhao, Lei, Zhao, Feng, Yates, Vincent, Pottier, Marc
• Primary Examiner(s): Knox, Tyler W
• Application Number:
  US15/365,715
• Publication Number:
  US 20180121847A1
• Time in Patent Office: 1,021 Days
• CPC Class Codes: G06Q 10/047 Optimisation of routes or paths, e.g. travelling salesman problem
• CPC Class Codes: G06Q 10/06311 Scheduling, planning or task assignment for a person or group
• CPC Class Codes: G06Q 10/06312 Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
• CPC Class Codes: G06Q 10/06313 Resource planning in a project environment
• CPC Class Codes: G06Q 10/06314 Calendaring for a resource
• CPC Class Codes: G06Q 10/06315 Needs-based resource requirements planning or analysis

Abstract

In some locations, such as airports, drivers that provide passenger transport must wait a significant distance away from the pickup location. To reduce the delay between a passenger requesting a ride and being picked up, drivers are pre-dispatched before being assigned to specific passengers. The demand for a future time period is predicted and the number of drivers that should be pre-dispatched to meet that demand is estimated. The estimated number of drivers is sent pre-dispatch instruction indicating they should leave the waiting area and head towards the pickup location.

Claims

1. A computer-implemented method for arranging transport services, the method comprising:

predicting a number of transport requests in a first geographic region during a future time period, the predicting including;

receiving one or more signals regarding a current time, wherein the one or more signals include data indicating a number of recent instances of a ride request application being opened determined based on messages transmitted by client mobile devices of one or more riders responsive to the ride request application being initiated on the client mobile device; and

applying a machine learning algorithm to the one or more signals to generate a prediction of the number of transport requests;

estimating unmet demand based on the predicted number of transport requests, the estimating comprising subtracting, from the predicted number of transport requests, a number of driver mobile computing devices which have previously received transmitted messages instructing an associated receiving driver to travel towards a specified location for which the receiving driver is still available for assignment;

receiving, from driver mobile computing devices over a computing network, data packets including GPS location data and availability data generated by the driver mobile computing devices, the GPS location data corresponding to positions of the driver mobile computing devices within or attached to driver vehicles at a current time;

identifying a plurality of the driver mobile computing devices based on the received GPS location data, each driver mobile computing device of the plurality of the driver mobile computing devices currently positioned in a second geographic region designated as a waiting area for the first geographic region and having a state, determined from the availability data, indicating availability of a corresponding driver to provide transport services;

determining a number of instances of a message to transmit to driver mobile computing devices, the message instructing a receiving driver to travel towards a specified location in the first geographic region, the number of instances determined based on the estimated unmet demand and a probability that, on receiving the message, the receiving driver will travel towards the specified location as instructed;

selecting a first set of driver mobile computing devices from the plurality of the driver mobile computing devices to message, the number of driver mobile computing devices in the first set being equal to the determined number of instances of the message;

transmitting, over the computing network to each driver mobile computing device of the first set of driver mobile computing devices, an instance of the message instructing the receiving driver to travel toward a specified location in the first geographic region without assigning a specific transport service or rider to the receiving driver; and

during a later time period;

receiving, from a client mobile device of a first rider, a transport request including a pickup location in the first geographic region;

receiving, from a second set of driver mobile computing devices from the plurality of the driver mobile computing devices over the computing network, data packets including updated GPS location data and availability data generated by the second set of driver mobile computing devices, the second set of driver mobile computing devices including at least some of the first set of driver mobile computing devices;

determining a subset of driver mobile computing devices from the second set of driver mobile computing devices responsive to the updated GPS location data indicating the corresponding driver has left the second geographic region and has moved more than a threshold distance along a route to the pickup location;

selecting, from the subset of driver mobile computing devices, a first driver mobile computing device based at least in part on a distance between the first driver mobile computing device and the pickup location, as indicated by the updated GPS location data;

transmitting, to the first driver mobile computing device, an invitation to provide the transport service for the first rider;

receiving, from the first driver mobile computing device, an indication of acceptance of the invitation;

assigning, responsive to the indication of acceptance, a driver corresponding to the first driver mobile computing device to provide the transport service for the first rider; and

transmitting a message to the first driver mobile computing device, the message causing the first driver mobile computing device to display a rider identity for the first rider, the pickup location, and a recommended route to the pickup location.

2. The method of claim 1, wherein the one or more signals further include an indication of a current time period, and the predicted number of transport requests is based on historical demand in time periods similar to the current time period.

3. The method of claim 2, wherein the current time period is at least one of:

a current time of day, a current day of the week, a current month, or a current date, and the time periods similar to the current period of time include at least one of;

a same time of day, a same day of the week, a same month, or a same date.

4. The method of claim 1, wherein the one or more signals further include at least one of:

flight data, location data received from passenger devices, data indicating a number of passenger devices recently switched on, calendars corresponding to passenger devices, or user profiles corresponding to passenger devices.

5. The method of claim 1, wherein the second geographic region is a subset of the first geographic region.

6. The method of claim 1, wherein selecting the first driver mobile computing device based at least in part on the distance between the first driver mobile computing device and the pickup location comprises selecting a driver mobile computing device of the subset of driver mobile computing devices that has a shortest estimated time of arrival at the pickup location.

7. The method of claim 1, wherein each driver mobile computing device in a first-in-first-out list has an individual probability of the corresponding driver responding to a message instructing the receiving driver to travel towards the specified location, and determining the number of instances of the message to transmit includes summing the individual probabilities until the sum exceeds the estimated unmet demand.

8. The method of claim 1, further comprising determining that a driver is still available for assignment based on GPS location data corresponding to the driver indicating the driver is traveling toward the pickup location.

19. The method of claim 1, wherein determining a subset of driver mobile computing devices from the second set of driver mobile computing devices is further responsive to the updated GPS location data indicating that the corresponding driver is on a road that is part of a reasonable route to the pickup location, the reasonable route being a primary route or one of a set of secondary routes generated by a routing algorithm.

9. A non-transitory computer-readable storage medium storing executable computer program code for arranging transport services in a first geographic region, the computer program code comprising instructions for:

predicting a number of transport requests in a first geographic region during a future time period, the predicting including;

receiving one or more signals regarding a current time, wherein the one or more signals include data indicating a number of recent instances of a ride request application being opened determined based on messages transmitted by client mobile devices of one or more riders responsive to the ride request application being initiated on the client mobile device; and

applying a machine learning algorithm to the one or more signals to generate a prediction of the number of transport requests;

estimating unmet demand based on the predicted number of transport requests, the estimating comprising subtracting, from the predicted number of transport requests, a number of driver mobile computing devices which have previously received transmitted messages instructing an associated receiving driver to travel towards a specified location for which the receiving driver is still available for assignment;

receiving, from driver mobile computing devices over a computing network, data packets including GPS location data and availability data generated by the driver mobile computing devices, the GPS location data corresponding to positions of the driver mobile computing devices within or attached to driver vehicles at a current time;

identifying a plurality of the driver mobile computing devices based on the received GPS location data, each driver mobile computing device of the plurality of the driver mobile computing devices currently positioned in a second geographic region designated as a waiting area for the first geographic region and having a state, determined from the availability data, indicating availability of a corresponding driver to provide transport services;

determining a number of instances of a message to transmit to driver mobile computing devices, the message instructing a receiving driver to travel towards a specified location in the first geographic region, the number of instances determined based on the estimated unmet demand and a probability that, on receiving the message, the receiving driver will travel towards the specified location as instructed;

selecting a first set of driver mobile computing devices from the plurality of the driver mobile computing devices to message, the number of driver mobile computing devices in the first set being equal to the determined number of instances of the message;

transmitting, over the computing network to each driver mobile computing device of the first set of driver mobile computing devices, an instance of the message instructing the receiving driver to travel toward a specified location in the first geographic region without assigning a specific transport service or rider to the receiving driver; and

during a later time period;

receiving, from a client mobile device of a first rider, a transport request including a pickup location in the first geographic region;

receiving, from a second set of driver mobile computing devices from the plurality of the driver mobile computing devices over the computing network, data packets including updated GPS location data and availability data generated by the second set of driver mobile computing devices, the second set of driver mobile computing devices including at least some of the first set of driver mobile computing devices;

determining a subset of driver mobile computing devices from the second set of driver mobile computing devices responsive to the updated GPS location data indicating the corresponding driver has left the second geographic region and has moved more than a threshold distance along a route to the pickup location;

selecting, from the subset of driver mobile computing devices, a first driver mobile computing device based at least in part on a distance between the first driver mobile computing device and the pickup location, as indicated by the updated GPS location data;

transmitting, to the first driver mobile computing device, an invitation to provide the transport service for the first rider;

receiving, from the first driver mobile computing device, an indication of acceptance of the invitation;

assigning, responsive to the indication of acceptance, a driver corresponding to the first driver mobile computing device to provide the transport service for the first rider; and

transmitting a message to the first driver mobile computing device, the message causing the first driver mobile computing device to display a rider identity for the first rider, the pickup location, and a recommended route to the pickup location.

10. The non-transitory computer-readable storage medium of claim 9, wherein the one or more signals further include an indication of a current time period, and the predicted number of transport requests is based on historical demand in time periods similar to the current time period.

11. The non-transitory computer-readable storage medium of claim 10, wherein the current time period is at least one of:

a current time of day, a current day of the week, a current month, or a current date, and the time periods similar to the current period of time include at least one of;

a same time of day, a same day of the week, a same month, or a same date.

12. The non-transitory computer-readable storage medium of claim 9, wherein the one or more signals further include at least one of:

flight data, location data received from passenger devices, data indicating a number of passenger devices recently switched on, calendars corresponding to passenger devices, or user profiles corresponding to passenger devices.

13. The non-transitory computer-readable storage medium of claim 9, wherein the second geographic region is a subset of the first geographic region.

14. The non-transitory computer-readable storage medium of claim 9, wherein the instructions for selecting the first driver mobile computing device based at least in part on the distance between the first driver mobile computing device and the pickup location comprise instructions for selecting a driver mobile computing device of the subset of driver mobile computing devices that has a shortest estimated time of arrival at the pickup location.

15. The non-transitory computer-readable storage medium of claim 9, wherein each driver mobile computing device in a first-in-first-out list has an individual probability of the receiving driver responding to a message instructing the corresponding driver to travel towards the specified location, and determining the number of instances of the message to transmit includes summing the individual probabilities until the sum exceeds the estimated unmet demand.

16. The non-transitory computer-readable storage medium of claim 9, wherein the instructions further comprise instructions for determining that a driver is still available for assignment based on GPS location data corresponding to the driver indicating the driver is traveling toward the pickup location.

17. A passenger transport system comprising:

a plurality of driver mobile computing devices, each driver mobile computing device located within or attached to a corresponding vehicle that is available to provide transport services, each driver mobile computing device including a display for presenting instructions to a driver; and

a matching server in communication with the plurality of driver mobile computing devices and the plurality of passenger devices over a computing network, the matching server configured to;

predict a number of transport requests in a first geographic region during a future time period, the predicting including;

receiving one or more signals regarding a current time, wherein the one or more signals include data indicating a number of recent instances of a ride request application being opened determined based on messages transmitted by client mobile devices of one or more riders responsive to the ride request application being initiated on the client mobile device; and

applying a machine learning algorithm to the one or more signals to generate a prediction of the number of transport requests;

estimate unmet demand based on the predicted number of transport requests, the estimating comprising subtracting, from the predicted number of transport requests, a number of driver mobile computing devices which have previously received transmitted messages instructing an associated receiving driver to travel towards a specified location for which the receiving driver is still available for assignment;

receive, from driver mobile computing devices over the computing network, data packets including GPS location data and availability data generated by the driver mobile computing devices, the GPS location data corresponding to positions of the driver mobile computing devices within or attached to driver vehicles at a current time;

identify a plurality of driver mobile computing devices based on the received GPS location data, each driver mobile computing device of the plurality of driver mobile computing devices currently positioned in a second geographic region designated as a waiting area for the first geographic region and having a state, determined from the availability data, indicating availability of a corresponding driver to provide transport services;

determine a number of instances of a message to transmit to driver mobile computing devices, the message instructing a receiving driver to travel towards a specified location in the first geographic region, the number of instances determined based on the estimated unmet demand and a probability that, on receiving the message, the receiving driver will travel towards the specified location as instructed;

select a first set of driver mobile computing devices from the plurality of the driver mobile computing devices to message, the number of driver mobile computing devices in the first set being equal to the determined number of instances of the message; and

transmit, over the computing network to each driver mobile computing device of the first set of driver mobile computing devices, an instance of the message instructing the receiving driver to travel toward a specified location in the first geographic region without assigning a specific transport service or rider to the receiving driver.

18. The passenger transport system of claim 17, wherein the matching server is further configured to:

during a later the given period of time;

receive, from a client mobile device of a first rider, a transport request including a pickup location in the first geographic region;

receive, from a second set of driver mobile computing devices from the plurality of the driver mobile computing devices over the computing network, data packets including updated GPS location data and availability data generated by the second set of driver mobile computing devices, the second set of driver mobile computing devices including at least some of the first set of driver mobile computing devices;

determine a subset of driver mobile computing devices from the second set of driver mobile computing devices responsive to the updated GPS location data indicating the corresponding driver has left the second geographic region and has moved more than a threshold distance along a route to the pickup location;

select, from the subset of driver mobile computing devices, a first driver mobile computing device based at least in part on a distance between the first driver mobile computing device and the pickup location, as indicated by the updated GPS location data;

transmit, to the first driver mobile computing device, an invitation to provide the transport service for the first rider;

receive, from the first driver mobile computing device, an indication of acceptance of the invitation;

assign, responsive to the indication of acceptance, a driver corresponding to the first driver mobile computing device to provide the transport service for the first rider; and

transmit a message to the first driver mobile computing device, the message causing the first driver mobile computing device to display a rider identity for the first rider, the pickup location, and a recommended route to the pickup location.