SYSTEM AND METHOD FOR PROCESSING SHIPMENT REQUESTS USING A MULTI-SERVICE SHIPPING PLATFORM
Systems and methods for processing shipment request by using a multi-carrier shipping services platform.
1 . A system comprising:
a set of APIs configured to:
receive a shipment request from a client computing system, wherein the shipment request comprises a set of time-in-transit requirements; and
provide a response to the shipment request to the client computing system;
a predictive model trained to predict a time-in-transit for each of a set of package percentiles, wherein the predictive model is trained on multi-service tracking data received from a set of shipping carriers; and
a request processor configured to:
generate the response using the set of time-in-transit requirements and the predictive model, wherein the response identifies at least one shipping carrier service.
2 . The system of claim 1 , wherein the system is a multi-tenant, multi-carrier shipping services platform.
3 . The system of claim 1 , wherein the set of time-in-transit requirements comprises a target time-in-transit for a target package percentile, wherein the at least one shipping carrier service identified in the response is predicted, by the predictive model, to deliver the target package percentile within the target time-in-transit.
4 . The system of claim 1 , wherein the multi-service tracking data is received from a scan event initiated by a computing device of a shipping carrier.
5 . The system of claim 1 , wherein the predictive model predicts the time-in-transit based on additional data, comprising at least one of weather data or traffic data.
6 . The system of claim 1 , wherein the predictive model predicts the time-in-transit based on data comprising a first location and a second location included in the shipment request.
7 . The system of claim 1 , wherein training the predictive model comprises:
determining a set of time intervals;
computing an actual delivered package percentile for each time interval of the set of time intervals; and
training the predictive model to predict the respective time interval for each of the resultant set of actual delivered package percentiles.
9 . The system of claim 1 , wherein the response is generated by aggregating a subset of the package percentiles meeting the time-in-transit requirements.
10 . The system of claim 1 , wherein the predictive model is trained in response to receipt of the shipment request.
11 . The system of claim 1 , wherein the identified shipping carrier service is used for delivering a parcel within the target time-in-transit for a package associated with the shipment request.
12 . A method comprising:
receiving a request from a client computing system, wherein the request comprises a target transit time for a target package;
generating a response to the request using a predictive model, trained to determine a delivery forecast for each of a set of package percentiles based on tracking data received from a set of shipping carriers, wherein the response identifies at least one shipping carrier from the set of carriers; and
providing the response to the client computing system.
13 . The method of claim 12 , wherein the predictive model is a machine learning model.
14 . The method of claim 12 , wherein the request further comprises a target percentage percentile, wherein the identified shipping carrier service is associated with a delivery forecast no slower than the target transit time for a package percentile no lower than the target package percentile.
15 . The method of claim 12 , wherein the predictive model further predicts the time-in-transit based on information included in the request.
16 . The method of claim 15 , wherein the information comprises geographic region information.
17 . The method of claim 12 , wherein the predictive model is trained responsive to receiving the request, wherein training the predictive model comprises:
determining training data, from the tracking data, that matches information from the request; and
training the predictive model using the training data.
18 . The method of claim 12 , wherein training the predictive model comprises generating time-in-transit data for each of a set of shipping lanes associated with the request by using the set of tracking data, wherein generating time-in-transit data for a shipping lane comprises aggregating data for historical data records from both directions for each of the set of shipping lanes, wherein the predictive model is trained on the time-in-transit data.
19 . The method of claim 12 , wherein the request is received from the client computing system via a public REST (Representational State Transfer) Application Programming Interface (API) of a shipping services platform.
20 . The method of claim 12 , wherein generating the response to the request using the target transit time and the predictive model comprises predicting a percentage of packages delivered within each of a set of potential transit times, and aggregating the percentiles to reach a threshold percentile.