IP Library Granted Patent US 12682379
Granted Patent B2
US 12682379 · App. 18/732,554 · Granted Jul 14, 2026

Customer preference driven vehicle supply management system

Inventors: Ram Kurapati (Southlake, TX); Munish Goyal (Yorktown Heights, NY); Muttu Vastrad (Frisco, TX); Audrey Mito (The Colony, TX); Smail Haddad (Plano, TX); Doug S. Campbell (McKinney, TX); Philip W. Ryan (Frisco, TX); Ramesh Gangadharan (Suwanee, GA); Mai Miller (Celina, TX); Srinivas Prasad (Plano, TX)
Assignees: TOYOTA MOTOR NORTH AMERICA, INC.; TOYOTA JIDOSHA KABUSHIKI KAISHA
G06Q30/0605G06Q10/087
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12682379
App. No.
18/732,554
Granted
Jul 14, 2026
Kind
B2
Abstract

A system includes an application configured to, based on user inputs, generate a target vehicle specification, and a database including a collection of inventory specifications of vehicles in a vehicle inventory. The system also includes a spec encoder configured to: encode the target vehicle specification as a first code; encode the collection of inventory specifications as a collection of second codes; and map the target vehicle specification and inventory specifications into a multidimensional space. The multidimensional space includes one dimension for each specification category. A preference class vector includes a weight for each dimension. A matching engine, based on the preference class vector, defines a search neighborhood around the target specification in the multidimensional space; identifies inventory specifications that fall within the search neighborhood, computes match scores for the identified vehicles; and reports the match scores and corresponding vehicles.

Claims (74)

1 . A system comprising:

a first processor operatively coupled to a memory comprising:

a user application configured to, based on inputs from a user, generate a target vehicle specification;

a second processor operatively coupled to a second memory comprising:

a database comprising a plurality of inventory specifications of a respective plurality of vehicles in a vehicle inventory;

a spec encoder configured to:

encode the target vehicle specification as a first code;

encode the plurality of inventory specifications as a plurality of respective second codes; and

map the target vehicle specification and the plurality of inventory specifications into a multidimensional specification space, wherein the multidimensional specification space includes one dimension for each specification category of the target vehicle specification;

a preference class vector comprising a respective weight for each dimension of the multidimensional specification space; and

a matching engine configured to:

based on the preference class vector, define a search neighborhood around the target vehicle specification in the multidimensional specification space;

identify inventory specifications of the plurality of inventory specifications that fall within the search neighborhood; and

for each identified inventory specification:

compute a match score; and

report, to the user, the match score and the vehicle corresponding to the identified inventory specification.

2 . The system of claim 1 , wherein the first code and the plurality of second codes are SR codes.

3 . The system of claim 1 , wherein the first code and the plurality of second codes are hexadecimal strings.

4 . The system of claim 1 , wherein the specification categories of the target vehicle specification include at least one of an exterior color, an interior color, an engine type, a transmission type, an accessory, or a trim package.

5 . The system of claim 1 , further comprising a user feedback module configured to:

receive user feedback regarding the vehicles corresponding to the identified inventory specifications; and

based on the user feedback, update the preference class vector.

6 . The system of claim 1 , further comprising a specification manager configured to, based on changes in the vehicle inventory, update the database comprising the plurality of inventory specifications.

7 . The system of claim 1 , further comprising a tightness updater configured to:

based on the mapped target vehicle specification, the mapped plurality of inventory specifications, and the search neighborhood, compute a tightness for the target vehicle specification; and

report the tightness to the user, or to a dealer or manufacturer.

8 . The system of claim 1 , further comprising gap glut analyzer configured to:

based on a plurality of target vehicle specifications and the plurality of inventory specifications, compute a gap-glut estimate; and

report the gap-glut estimate to a dealer or manufacturer.

9 . The system of claim 8 , further comprising a goal maximizer configured to, based on the gap-glut estimate:

request changes to the vehicle inventory; or

adjust prices of at least some vehicles in the vehicle inventory.

10 . The system of claim 1 , wherein the user application is a smartphone application or web application.

11 . A non-transitory computer-readable medium containing instructions which, when executed by at least one processor, perform operations which include:

with a user application executing on a first processor, based on inputs from a user, generating a target vehicle specification;

with a database stored in a memory of a second processor, storing a plurality of inventory specifications of a respective plurality of vehicles in a vehicle inventory;

with a spec encoder executing on the second processor:

encoding the target vehicle specification as a first code;

encoding the plurality of inventory specifications as a plurality of respective second codes; and

mapping the target vehicle specification and the plurality of inventory specifications into a multidimensional specification space, wherein the multidimensional specification space includes one dimension for each specification category of the target vehicle specification;

with a preference class vector stored in the memory, storing a respective weight for each dimension of the multidimensional specification space;

with a matching engine:

based on the preference class vector, defining a search neighborhood around the target vehicle specification in the multidimensional specification space;

identifying inventory specifications of the plurality of inventory specifications that fall within the search neighborhood; and

for each identified inventory specification:

computing a match score; and

reporting, to the user, the match score and the vehicle corresponding to the identified inventory specification.

12 . The non-transitory computer-readable medium of claim 11 , wherein the first code and the plurality of second codes are SR codes.

13 . The non-transitory computer-readable medium of claim 11 , wherein the first code and the plurality of second codes are hexadecimal strings.

14 . The non-transitory computer-readable medium of claim 11 , wherein the specification categories of the target vehicle specification include at least one of an exterior color, an interior color, an engine type, a transmission type, an accessory, or a trim package.

15 . The non-transitory computer-readable medium of claim 11 , wherein the instructions further comprise, with a user feedback module:

receiving user feedback regarding the vehicles corresponding to the identified inventory specifications; and

based on the user feedback, updating the preference class vector.

16 . The non-transitory computer-readable medium of claim 11 , wherein the instructions further comprise, with a specification manager, based on changes in the vehicle inventory, updating the database comprising the plurality of inventory specifications.

17 . The non-transitory computer-readable medium of claim 11 , wherein the instructions further comprise, with a tightness updater:

based on the mapped target vehicle specification, the mapped plurality of inventory specifications, and the search neighborhood, computing a tightness for the target vehicle specification; and

reporting the tightness to the user, or to a dealer or manufacturer.

18 . The non-transitory computer-readable medium of claim 11 , wherein the instructions further comprise, with a gap glut analyzer:

based on a plurality of target vehicle specifications and the plurality of inventory specifications, computing a gap-glut estimate; and

reporting the gap-glut estimate to a dealer or manufacturer.

19 . The non-transitory computer-readable medium of claim 18 , wherein the instructions further comprise, with a goal maximizer, based on the gap-glut estimate:

requesting changes to the vehicle inventory; or

adjusting prices of at least some vehicles in the vehicle inventory.

20 . A computer implemented method, comprising, in real time, with a first processor comprising a memory:

defining a vehicle specification format comprising a plurality of specification categories;

defining a specification code format comprising at least one digit for each specification category of the plurality of specification categories;

receiving, from a second processor, a vehicle specification in the vehicle specification format;

mapping the vehicle specification into a multidimensional specification space, wherein the multidimensional specification space includes one dimension for each specification category of the vehicle specification;

identifying a second vehicle specification proximate to the vehicle specification in the multidimensional space;

mapping the second vehicle specification into the specification code format;

representing the mapped second vehicle specification as a two-dimensional SR code;

storing the two-dimensional SR code;

receiving a user request; and

based on the user request, electronically transmitting the two-dimensional SR code to the second processor.