Apparatus and method for predictive dispatch for geographically distributed, on-demand services

• US 10,127,519 B2
• Filed: 10/22/2016
• Issued: 11/13/2018
• Est. Priority Date: 10/23/2015
• Status: Active Grant

• Current Assignee: Matthew Walton
• Original Assignee: Prahfit Inc.
• Inventors: Walton, Matthew
• Primary Examiner(s): Glass, Russell S
• Application Number:
  US15/331,842
• Publication Number:
  US 20170116566A1
• Time in Patent Office: 752 Days
• Field of Search: 705 28
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/083 Shipping
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders
• CPC Class Codes: G06Q 30/0283 Price estimation or determination
• CPC Class Codes: G06Q 30/0284 Time or distance, e.g. usage of parking meters or taximeters

Abstract

Provided is a process of coordinating dynamic on-demand logistics to deliver goods or services between geographically distributed heterogeneous sources of inventory and geographically distributed consumers at scales exceeding metropolitan areas of more than one million residents, with more than 10,000 users, more than 1,000 vehicle operators, and more than 10 geographically distributed sources of inventory.

Claims

1. A method of coordinating dynamic on-demand logistics to deliver goods or services between geographically distributed heterogeneous sources of inventory and geographically distributed consumers at scales exceeding metropolitan areas of more than one million residents, with more than 10,000 users, more than 1,000 vehicle operators, and more than 10 geographically distributed sources of inventory, the method comprising:

obtaining, with one or more processors, an order, the order indicating goods or services to be provided, and indicating a destination geolocation at which the goods or services are to be provided;

identifying, with one or more processors, a plurality of candidate sources of inventory based on proximity to the destination geolocation;

searching, with one or more processors, inventories of the candidate sources to determine qualified candidate sources in response to determining those qualified candidate sources have in inventory items needed to provision the goods or services specified by the order;

identifying, with one or more processors, a one or more geographic zones based on those geographic zones containing at least one of the qualified candidate sources, at least some of the geographic zones having a plurality of qualified candidate sources;

for each of the geographic zones;

obtaining, with one or more processors, real-time geolocations of delivery vehicles within the respective geographic zone;

obtaining, with one or more processors, real-time time estimates for delivery by the respective delivery vehicles;

obtaining, with one or more processors, respective geolocations of qualified candidate sources within the respective geographic zone;

obtaining, with one or more processors, respective amounts of inventory for the qualified candidate sources within the respective geographic zone; and

obtaining, with one or more processors, other orders for which the respective geographic zone contains a qualified candidate source; and

determining, with one or more processors, a delivery plan for the respective geographic zone based on the obtained real-time geolocations of the delivery vehicles, real-time time estimates for delivery, geolocations of the qualified candidate sources, amounts of inventory, and other orders, wherein determining a delivery plan comprises;

determining a cost score for each of a plurality of candidate delivery scenarios, each delivery scenario including a given delivery vehicle and a given qualified candidate source;

determining that the candidate delivery scenarios do not exceed the amounts of inventory; and

selecting a plurality of delivery scenarios for the order and the other orders that reduces an aggregate cost score for the respective geographic zone; and

sending, with one or more processors, at least one delivery plan to a mobile computing device of an operator of a delivery vehicle specified by the at least one delivery plan.

2. The method of claim 1, comprising:

determining that more than a threshold amount of time has elapsed since sending the at least one delivery plan to the operator of the delivery vehicles specified by the at least one delivery plan and, in response, adding the order to another queue to be re-assigned.

3. The method of claim 1, comprising:

determining that the order includes multiple items that are individually or collectively larger than a threshold weight or volume and, in response, when determining a delivery plan, determining that the candidate delivery scenarios include a delivery vehicle capable of transporting the multiple items.

4. The method of claim 1, comprising:

determining that the order includes multiple items that are individually or collectively larger than a threshold weight or volume and, in response, sending the order to a courier service.

5. The method of claim 1, comprising:

ranking delivery scenarios based on cost scores;

identifying a plurality delivery scenarios satisfying a threshold ranking;

designating in memory inventory corresponding to the identified delivery scenarios as pending.

6. The method of claim 5, wherein the threshold ranking is the top three delivery scenarios having the three lowest cost scores.

7. The method of claim 1, comprising:

repeatedly optimizing delivery scenarios in each geographic zone more often than once per hour, each optimizing including more than 10 orders, more than 100 delivery vehicles, and more than 5 sources of inventory.

8. The method of claim 1, comprising:

receiving feedback from consumers, candidate sources, and operators of delivery vehicles; and

adjusting parameters in an algorithm by which delivery plans are selected based on the feedback.

9. The method of claim 1, comprising:

determining aggregate measures of demand from a plurality of the geographic zones; and

sending the aggregate measures of demand to mobile computing devices of operators of delivery vehicles to rebalance the location of delivery vehicles among the geographic zones.

10. The method of claim 1, wherein:

determining a delivery plan is based on a customer profile of a customer specifying the order.

11. The method of claim 1, wherein:

determining a delivery plan is based on an estimated amount of revenue for each delivery scenario.

12. The method of claim 1, wherein:

determining a delivery plan is based on discounts associated with at least some delivery scenarios.

13. The method of claim 1, wherein:

determining a delivery plan is based on estimated traffic delays associated with at least some delivery scenarios.

14. The method of claim 1, wherein:

determining a delivery plan is based on an age of the order.

15. The method of claim 1, wherein:

determining a delivery plan is based on a priority of the order.

16. The method of claim 1, wherein:

determining a delivery plan is based on a simulated annealing algorithm that identifies a minimum or maximum of a simulated energy landscape corresponding to aggregate cost score for the respective geographic zone.

17. The method of claim 1, comprising:

receiving another order from a client computing device;

determining that there are no qualified candidate sources within a threshold distance to a destination geolocation of the other order; and

sending the client computing device instructions to display a user interface with an input by which a user inputs a request to increase the threshold distance.

18. The method of claim 1, wherein:

determining a delivery plan is based on a gradient descent of the aggregate cost score for the respective geographic zone.

19. The method of claim 1, wherein:

determining a delivery plan comprises steps for optimizing delivery scenarios within a geographic zone.

20. A system, comprising:

one or more processors; and

memory storing instructions that when executed by at least some of the processors effectuate operations comprising;

obtaining an order, the order indicating goods or services to be provided, and indicating a destination geolocation at which the goods or services are to be provided;

identifying a plurality of candidate sources of inventory based on proximity to the destination geolocation;

searching inventories of the candidate sources to determine qualified candidate sources in response to determining those qualified candidate sources have in inventory items needed to provision the goods or services specified by the order;

identifying a one or more geographic zones based on those geographic zones containing at least one of the qualified candidate sources, at least some of the geographic zones having a plurality of qualified candidate sources;

for each of the geographic zones;

obtaining real-time geolocations of delivery vehicles within the respective geographic zone;

obtaining real-time time estimates for delivery by the respective delivery vehicles;

obtaining respective geolocations of qualified candidate sources within the respective geographic zone;

obtaining respective amounts of inventory for the qualified candidate sources within the respective geographic zone; and

obtaining other orders for which the respective geographic zone contains a qualified candidate source; and

determining a delivery plan for the respective geographic zone based on the obtained real-time geolocations of the delivery vehicles, real-time time estimates for delivery, geolocations of the qualified candidate sources, amounts of inventory, and other orders, wherein determining a delivery plan comprises;

determining a cost score for each of a plurality of candidate delivery scenarios, each delivery scenario including a given delivery vehicle and a given qualified candidate source;

determining that the candidate delivery scenarios do not exceed the amounts of inventory; and

selecting a plurality of delivery scenarios for the order and the other orders that reduces an aggregate cost score for the respective geographic zone; and

sending at least one delivery plan to a mobile computing device of an operator of a delivery vehicle specified by the at least one delivery plan.