Predictive commodity classification
AssigneeUnited Parcel Service
InventorsDale A. Black, Barry Bramble
Life of this patent
Estimating the term…
Abstract
Systems and methods for predicting commodity classifications for a shipment that increase the efficiency of clearing the shipment through a customs authority. Such systems and methods may utilize machine learning techniques to improve the accuracy of the predicted classifications.
Claims20
A computer-implemented method for predicting commodity classifications for a shipment, the method comprising:
- receiving an indication that delivery of the shipment has been halted at a location associated with a remote computing device associated with a customs authority;
- responsive to receiving the indication, generating, by a computing device, a commercial invoice for the shipment, wherein the commercial invoice includes a listing of commodities included in the shipment;
- predicting classifications, by the computing device, for each of the commodities listed on the commercial invoice using a knowledgebase data repository, wherein each of the predicted classifications has a confidence score;
- generating, by the computing device, a classified invoice based upon the commercial invoice and the predicted classifications;
- based upon the confidence score for each of the predicted classifications exceeding a threshold level, automatically generating, without additional interaction by a user, by the computing device, a digital customs entry having data fields for an accounting of duties, taxes and fees for the shipment based upon the classified invoice;
- automatically communicating, by the computing device, the digital customs entry to the remote computing device associated with the customs authority;
- receiving, with the computing device, a customs clearance from the customs authority in response to the communicated digital customs entry;
- responsive to receiving the customs clearance from the customs authority, causing the remote computing device associated with the customs authority to release the shipment from the location;
- and
- based upon receiving the customs clearance, automatically updating, by the computing device, the knowledgebase data repository.
A non-transitory computer-readable medium storing computer-usable instructions that, when used by one or more computing devices, cause the one or more computing devices to perform operations comprising:
- receiving an indication that delivery of the shipment has been halted at a location associated with a remote computing device associated with a customs authority;
- responsive to receiving the indication, generating a commercial invoice for the shipment, wherein the commercial invoice includes a listing of commodities included in the shipment;
- predicting classifications for each of the commodities listed on the commercial invoice using a knowledgebase data repository, wherein each of the predicted classifications has a confidence score, wherein the confidence score is calculated based on at least one of a part number, a country of origin, a country of importation, a part description, licenses, certificates, and a comparison of an importer or exporter with a consignee or shipper;
- generating a classified invoice based upon the commercial invoice and the predicted classifications;
- based upon the confidence score for each of the predicted classifications exceeding a threshold level, automatically generating, without additional interaction by a user, a digital customs entry having data fields for an accounting of duties, taxes and fees for the shipment based upon the classified invoice;
- automatically, communicating the digital customs entry to a remote computing device associated with the customs authority;
- receiving a customs clearance from the customs authority in response to the communicated digital customs entry;
- responsive to receiving the customs clearance from the customs authority, causing the remote computing device associated with the customs authority to release the shipment from the location;
- and
- based upon receiving the customs clearance, automatically updating the knowledgebase data repository via machine learning techniques.
A non-transitory computer-readable medium storing computer-usable instructions that, when used by one or more computing devices, cause the one or more computing devices to perform operations comprising:
- receiving an indication that delivery of the shipment has been halted at a location associated with a remote computing device associated with a customs authority;
- responsive to receiving the indication, generating a commercial invoice for the shipment, wherein the commercial invoice includes a listing of commodities included in the shipment;
- predicting classifications for each of the commodities listed on the commercial invoice using a knowledgebase data repository, wherein each of the predicted classifications has a confidence score, wherein the confidence score is calculated based on at least one of a part number, a country of origin, a country of importation, a part description, licenses, certificates, and a comparison of an importer or exporter with a consignee or shipper;
- generating a classified invoice based upon the commercial invoice and the predicted classifications;
- based upon the confidence score for each of the predicted classifications exceeding a threshold level, automatically generating, without additional interaction by a user, by the computing device, a digital customs entry having data fields for an accounting of duties, taxes and fees for the shipment based upon the classified invoice;
- automatically, communicating the digital customs entry to a remote computing device associated with the customs authority;
- receiving a customs clearance from the customs authority in response to the communicated digital customs entry;
- responsive to receiving the customs clearance from the customs authority, causing the remote computing device associated with the customs authority to release from the location;
- and
- based upon receiving the customs clearance, automatically updating the knowledgebase data repository via machine learning techniques, including increasing or maintaining the confidence score for each of the predicted classifications that received the customs clearance.
Dependencies are inferred from the claim text.