Neural networks to generate reliability scores
AssigneeAmazon Technologies, Inc.
InventorsHaroon Ali Akbar, Juan Pablo Padilla, Rahul Pruthi
Life of this patent
Estimating the term…
Abstract
Systems and methods are described relating to score generation for an account of a delivery service using one or more neural networks. The one or more neural networks can generate the score using time series data including metrics associated with the account. In response to a time slot being selected by the account from the set of time slots, the score can be used to generate safeguards for the account, where the safeguards are to be applied prior to execution of one or more scheduled deliveries associated with the time slot.
Claims21
1-20. (canceled)
A computer-implemented method, comprising:
- generating a plurality of time periods that correspond to one or more scheduled deliveries from a station;
- obtaining one or more metrics of a carrier associated with a delivery service, the one or more metrics recorded over the plurality of time periods;
- generating a quality score of the carrier based, at least in part, on the one or more metrics recorded over the plurality of time periods;
- upon determining that the quality score of the carrier is above a threshold, transmitting an order to a client device associated with the carrier;
- receiving a selection of a time period of the plurality of time periods;
- determining that the client device associated with the carrier is located in or proximate to a geographical region that includes the station;
- identifying one or more instructions to be applied to the carrier associated with the selected time period based, at least in part, on the quality score of the carrier;
- and
- causing the client device to indicate the one or more instructions that are applied to the carrier.
A computer-implemented method, comprising:
- obtaining multiple ratings of a carrier associated with a delivery service recorded over a plurality of time periods;
- generating an overall rating of the carrier based, at least in part, on the multiple ratings of the carrier recorded over the plurality of time periods;
- upon determining that the overall rating of the carrier is above a threshold, transmitting an offer for a scheduled delivery to a client device associated with the carrier, the offer not available to other carriers having an overall rating below the threshold;
- receiving a selection of a time period of a plurality of time periods that correspond to one or more scheduled deliveries;
- identifying one or more instructions to be applied to the carrier and associated with the selected time period based, at least in part, on the overall rating of the carrier;
- and
- causing the client device to indicate the one or more instructions that are applied to the carrier.
A computer-implemented method, comprising:
- generating a plurality of time periods that correspond to one or more scheduled deliveries from a station;
- receiving a selection of a time period of the plurality of time periods;
- determining that a client device associated with a carrier of a delivery service is located in proximity to a geographical region that includes the station;
- obtaining one or more ratings of the carrier recorded over a plurality of time periods;
- generating an overall rating of the carrier based, at least in part, on the one or more ratings recorded over the plurality of time periods;
- upon determining that the overall rating of the carrier is above a threshold and the carrier is located in proximity to the geographical region that includes the station, transmitting an order to the client device associated with the carrier;
- identifying one or more instructions to be applied to the carrier and associated with the selected time period based, at least in part, on the overall rating of the carrier;
- and
- causing the client device to indicate the one or more instructions that are applied to the carrier.
Dependencies are inferred from the claim text.