# US 12,112,294 B2 · Multi-phase consolidation optimization tool

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,112,294 B2
- Document ID: US12112294B2
- Status: Granted
- Kind: B2
- Application number: 18/115,422
- Filed: 2023-02-28
- Published: 2023-07-06
- Granted: 2024-10-08
- Earliest claimed: 2019-08-13
- Assignees: United Parcel Service of America, Inc.
- Applicants: United Parcel Service Of America, Inc.
- Inventors: Benjamin Grisz; Casey Mccord; Emily Richards; Carol Crawford; Jose Landeros; Ayan Das; Christopher Martin Rubio
- Web page: https://patentgrep.com/patent/US12112294B2

## CPC

- G06Q 10/0838: {Historical data}
- G06F 8/38: for implementing user interfaces
- G06F 9/451: Execution arrangements for user interfaces
- G06N 3/042: Knowledge-based neural networks; Logical representations of neural networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/09: Supervised learning
- G06N 20/00: Machine learning
- G06Q 10/083: Shipping
- G06Q 10/0831: Overseas transactions
- G06Q 10/087: Inventory or stock management, e.g. order filling, procurement or balancing against orders
- G06Q 10/10: Office automation; Time management
- G06Q 30/0635: {replenishment orders; recurring orders}
- G06Q 50/26: Government or public services (business processes related to the transportation industry G06Q50/40)
- G06N 3/045: Combinations of networks
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06N 20/20: Ensemble learning
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Rule data sets are received. These rule sets are associated with constraints controlling how records that are associated with the goods are consolidated. These goods are to be received for importing. An estimate score indicative of the risk for inspection for a first set of goods that are to be imported is generated. Based at least in part on the rule data sets and the generated estimate, a plurality of records are consolidated to a single instance for the first set of goods. Based on the consolidating, a user interface is caused to be generated that renders information associated with the consolidating.

## Claims

### Claim 1

A method comprising:
training a machine learning model by:
extracting a set of historical features from historical data comprising past purchase orders that have been labeled as inspected or not inspected;
converting the set of historical features into one or more representative feature vectors; and
processing the one or more representative feature vectors using the machine learning model to identify a vector space representative of the set of historical features; subsequent to training the machine learning model:
extracting, by at least one computer processor, a set of features from a plurality of purchase orders, wherein each purchase order of the plurality of purchase orders comprises a feature of a good that is to be received for importing;
converting, by the at least one computer processor, the set of features into one or more feature vectors;
processing, by the at least one computer processor, the one or more feature vectors using the machine learning model to generate a prediction for a set of goods associated with the plurality of purchase orders of being inspected by a customs clearance agent, wherein the machine learning model generates the prediction based at least in part on determining a distance, in the vector space, between the one or more feature vectors and the one or more representative feature vectors;
accessing rule data sets associated with constraints controlling how the plurality of purchase orders is to be consolidated to an entry, wherein the entry corresponds to a customs clearance document identifying at least one of the set of goods that is to be received for importing;
based at least in part on the prediction and the rule data sets, automatically generating, by the at least one computer processor, an estimated risk for inspection by the customs clearance agent for the set of goods;
based at least in part on the estimated risk, automatically consolidating, by the at least one computer processor, the plurality of purchase orders to a single entry for importing the set of goods;
in response to automatically consolidating the plurality of purchase orders, storing, by the at least one computer processor, the single entry to a storage device such that the plurality of purchase orders is configured to be accessed from the storage device via a single computer input/output operation; and
based at least in part on consolidating the plurality of purchase orders, causing a user interface to be generated that renders information associated with consolidating the plurality of purchase orders.

### Claim 2 (depends on 1)

The method of claim 1, wherein the rule data sets include one or more rules from a customs agency.

### Claim 3 (depends on 2)

The method of claim 2, wherein the rule data sets include a predefined rule that the customs agency enforces and is to be abided by, and the one or more rules include at least one of a first rule that a first specific good is always inspected regardless of where the first specific good came from or a second rule that a second specific good is only inspected if the second specific good came from a specific geographic location associated with pestilence or disease.

### Claim 4 (depends on 1)

The method of claim 1, wherein the rule data sets include one or more rules from a buyer associated with the plurality of purchase orders.

### Claim 5 (depends on 4)

The method of claim 4, wherein the one or more rules includes a predefined rule that the buyer enforces and is to be abided by, and the one or more rules include at least one of a first rule that indicates that first certain goods are not to be contained in a same entry for importing or a second rule that second certain goods are always to be contained in the same entry for importing.

### Claim 6 (depends on 1)

The method of claim 1, further comprising computing, by the at least one computer processor, a cost savings based on consolidating the plurality of purchase orders to the single entry, wherein the cost savings is included in the user interface.

### Claim 7 (depends on 1)

The method of claim 1, wherein consolidating the plurality of purchase orders is further based at least in part on sensor data obtain from a freight vessel.

### Claim 8

One or more non-transitory computer storage media having computer-executable instructions embodied thereon that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
subsequent to training a machine learning model:
extracting a set of features from a plurality of purchase orders, wherein each purchase order of the plurality of purchase orders comprises a feature of a good that is to be received for importing;
converting the set of features into one or more feature vectors;
processing the one or more feature vectors using the machine learning model to generate a prediction for a set of goods associated with the plurality of purchase orders of being inspected by a customs clearance agent, wherein the machine learning model generates the prediction based at least in part on determining a distance, in a vector space, between the one or more feature vectors and one or more representative feature vectors, and training the machine learning model comprises:
extracting a set of historical features from historical data comprising past purchase orders that have been labeled as inspected or not inspected;
converting the set of historical features into the one or more representative feature vectors; and
processing the one or more representative feature vectors using the machine learning model to identify the vector space representative of the set of historical features;
accessing at least one rule data set associated with constraints controlling how the plurality of purchase orders is to be consolidated to an entry, wherein the entry corresponds to a customs clearance document identifying at least one of the goods that is to be received for importing;
based at least in part on the prediction and the at least one rule data set, automatically generating, for each good of the set of goods, an estimated risk for inspection by the customs clearance agent;
based at least in part on the estimated risk for each good of the set of goods, automatically consolidating the plurality of purchase orders to a single entry for importing the set of goods;
in response to automatically consolidating the plurality of purchase orders, storing the single entry to a storage device; and
based at least in part on consolidating the plurality of purchase orders, causing a user interface to be generated that renders information associated with consolidating the plurality of purchase orders.

### Claim 9 (depends on 8)

The one or more non-transitory computer storage media of claim 8, wherein the at least one rule data set includes one or more rules from a customs agency.

### Claim 10 (depends on 8)

The one or more non-transitory computer storage media of claim 8, wherein the at least one rule data set includes one or more rules from a buyer or a seller associated with the plurality of purchase orders.

### Claim 11 (depends on 8)

The one or more non-transitory computer storage media of claim 8, wherein the operations further comprise computing a cost savings based on consolidating the plurality of purchase orders to the single entry, and the cost savings is included in the user interface.

### Claim 12 (depends on 8)

The one or more non-transitory computer storage media of claim 8, wherein consolidating the plurality of purchase orders is further based at least in part on sensor data obtain from a freight vessel.

### Claim 13

A system comprising:
one or more processors; and
one or more computer storage media storing computer-useable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
training a machine learning model by:
extracting a set of historical features from historical data comprising past purchase orders that have been labeled as inspected or not inspected;
converting the set of historical features into one or more representative feature vectors; and
processing the one or more representative feature vectors using the machine learning model to identify a vector space representative of the set of historical features; subsequent to training the machine learning model:
extracting, by at least one computer processor, a set of features from a plurality of purchase orders, wherein each purchase order of the plurality of purchase orders comprises a feature of a good that is to be received for importing;
converting, by the at least one computer processor, the set of features into one or more feature vectors;
processing, by the at least one computer processor, the one or more feature vectors using the machine learning model to generate a prediction for each good of a set of goods associated with the plurality of purchase orders of being inspected by a customs clearance agent, wherein the machine learning model generates the prediction based at least in part on determining a distance, in the vector space, between the one or more feature vectors and the one or more representative feature vectors;
accessing at least one rule data set associated with constraints controlling how the plurality of purchase orders is to be consolidated to an entry, wherein the entry correspond to a customs clearance document identifying at least one of the goods that is to be received for importing;
based at least in part on the prediction and the at least one rule data set, generating, for each good of the set of goods, an estimated risk for inspection by the customs clearance agent;
based at least in part on the estimated risk for each good of the set of goods, automatically consolidating the plurality of purchase orders to a single entry for importing the set of goods;
storing the single entry to a storage device; and
based at least in part on consolidating the plurality of purchase orders, causing a user interface to be generated that renders information associated with consolidating the plurality of purchase orders.

### Claim 14 (depends on 13)

The system of claim 13, wherein the at least one rule data set includes one or more rules from a buyer or a seller associated with the plurality of purchase orders.

### Claim 15 (depends on 14)

The system of claim 14, wherein the one or more rules include a predefined rule that the seller or the buyer enforces and is to be abided by, and the one or more rules include at least one of a first rule that indicates that first certain goods are not to be contained in a same entry for importing or a second rule that second certain goods are always to be contained in the same entry for importing.

### Claim 16 (depends on 13)

The system of claim 13, wherein the at least one rule data set includes one or more rules from a customs agency.

### Claim 17 (depends on 16)

The system of claim 16, wherein the at least one rule data set includes a predefined rule that the customs agency enforces and is to be abided by, and the one or more rules includes at least one of a first rule that a first specific good is always inspected regardless of where the first specific good came from or a second rule that a second specific good is only inspected if the second specific good came from a specific geographic location associated with pestilence or disease.

### Claim 18 (depends on 13)

The system of claim 13, wherein the operations further comprise computing a cost savings based on consolidating the plurality of purchase orders to the single entry, and the cost savings is included in the user interface.