System for calibrating and validating parameters for optimization

Patent metadata

Field	Value
Publication number	US-9705751-B1
Application number	US-201615335070-A
Country	US
Kind code	B1
Filing date	Oct 26, 2016
Priority date	Mar 31, 2016
Publication date	Jul 11, 2017
Grant date	Jul 11, 2017

Abstract

A computing device quantifies an expected benefit from a calibrated coefficient of variation (CV) and/or a calibrated service level (SL). The target optimization model determines a number and a time a new requisition is placed for an item at each node of the plurality of nodes. A validation time value is updated using an incremental time value and the process is repeated until the validation time value is greater than or equal to a stop time.

First claim

What is claimed is:

1. A non-transitory computer-readable medium having stored thereon computer-readable instructions that when executed by a computing device cause the computing device to:
automatically select a calibrated parameter value of a first parameter;
receive an indicator of a validation horizon time period, wherein the validation horizon time period includes a start time, a stop time, and an incremental time;
automatically initialize a validation time value based on the start time;
automatically read requisition history data from a requisition history dataset, wherein the requisition history data includes previous requisitions placed during a time period prior to the start time for an item in a network that includes a plurality of nodes;
(a) automatically generate demand data for each node of the plurality of nodes using a forecast model with the requisition history data and the selected calibrated parameter value, wherein the forecast model is configured to forecast a demand associated with the item at each node of the plurality of nodes;
(b) automatically update the requisition history data and compute a simulated key performance indicator (KPI) value of a KPI by executing a target optimization model with the generated demand data, wherein the target optimization model is configured to determine a number and a time a new requisition is placed for the item at each node of the plurality of nodes;
(c) automatically store the computed, simulated KPI value in association with the selected initial validation time value;
(d) automatically update the initialized validation time value using the incremental time;
automatically repeat (a)-(d) until the updated, initialized validation time value is greater than or equal to the stop time;
automatically compute an aggregated KPI value as a sum of the stored KPI values for each node at each value of the initialized validation time value;
output a comparison between the computed, aggregated KPI values and historical KPI data computed from an actual requisition history for each node of the plurality of nodes during the validation horizon time period; and
automatically optimize a stockpile of the item in the network for each node of the plurality of nodes using the selected calibrated parameter value.

2. The non-transitory computer-readable medium of claim 1 , wherein the KPI is at least one of an on-hand disbursement value, a backlog value, a ready rate value, a fill rate value, and a back order ratio value.

3. The non-transitory computer-readable medium of claim 1 , wherein the first parameter is at least one of a service level and a coefficient of variation.

4. The non-transitory computer-readable medium of claim 1 , wherein the computer-readable instructions further cause the computing device to automatically output a comparison between the computed, aggregated KPI values and historical KPI data computed from an actual requisition history for each node of the plurality of nodes during the validation horizon time period to verify that the selected calibrated parameter value is configured correctly.

5. The non-transitory computer-readable medium of claim 1 , wherein selecting the calibrated parameter value of the first parameter comprises computer-readable instructions that further cause the computing device to:
receive indicators of a start value, a stop value, and an incremental value of the first parameter;
receive an indicator of a calibration horizon time period, wherein the calibration horizon time period includes a calibration start time, a calibration stop time, and a calibration incremental time;
automatically initialize a calibration time value based on the calibration start time;
automatically select a test value for the first parameter as the start value;
automatically read second requisition history data from the requisition history dataset, wherein the second requisition history data includes previous requisitions placed during the time period prior to the calibration start time for the item in the network that includes the plurality of nodes;
(e) automatically generate second demand data for each node of the plurality of nodes using the forecast model with the second requisition history data and the selected test value;
(f) automatically update the second requisition history data and compute a second simulated KPI value of the KPI by executing the target optimization model with the generated second demand data;
(g) automatically store the computed, second simulated KPI value in association with the selected test value and the initialized calibration time value
(h) automatically update the initialized calibration time value using the calibration incremental time; and
(i) automatically repeat (e)-(h) until the updated, initialized calibration time value is greater than or equal to the calibration stop time.

6. The non-transitory computer-readable medium of claim 5 , wherein after (i), selecting the calibrated parameter value of the first parameter further comprises computer-readable instructions that further cause the computing device to:
(j) automatically compute a second aggregated KPI value as a sum of the stored second KPI values for each node at each value of the initialized calibration time value;
(k) automatically update the selected test value using the incremental value;
(l) automatically repeat (e)-(k) until the updated, selected test value is greater than or equal to the stop value; and
automatically execute a stockpile parameter optimization model with the computed, second aggregated KPI values and the computed second aggregated KPI value;
wherein the calibrated parameter value of the first parameter is selected based on results of execution of the stockpile parameter optimization model that ranks each value of the selected test value.

7. The non-transitory computer-readable medium of claim 5 , wherein the computer-readable instructions further cause the computing device to automatically output the selected calibrated parameter value.

8. The non-transitory computer-readable medium of claim 5 , wherein the computer-readable instructions further cause the computing device to automatically output a comparison between the computed, second aggregated KPI values and historical KPI data computed from an actual requisition history for each node of the plurality of nodes during the calibration horizon time period.

9. The non-transitory computer-readable medium of claim 6 , wherein the stockpile parameter optimization model minimizes a total network backorder while satisfying a budget constraint defined by data read from a stockpile budget dataset.

10. The non-transitory computer-readable medium of claim 6 , wherein the stockpile parameter optimization model maximizes a total network service level while satisfying a budget constraint defined by data read from a stockpile budget dataset.

11. The non-transitory computer-readable medium of claim 1 , wherein the automatically selecting the calibrated parameter value of the first parameter and the automatically computing the aggregated KPI value are performed periodically.

12. A computing device comprising:
a processor; and
a non-transitory computer-readable medium operably coupled to the processor, the computer-readable medium having computer-readable instructions stored thereon that, when executed by the processor, cause the computing device to
automatically select a calibrated parameter value of a first parameter;
receive an indicator of a validation horizon time period, wherein the validation horizon time period includes a start time, a stop time, and an incremental time;
automatically initialize a validation time value based on the s

Assignees

• Sas Inst Inc

Inventors

• Yi Jinxin
• Kacar Necip Baris
• Valsaraj Varunraj

Classifications

• H04L41/0823

  characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability (for optimising operational conditions of wireless networks H04W24/02) · CPC title

• H04L41/145

  involving simulating, designing, planning or modelling of a network · CPC title

• G06F11/00 Primary

  Error detection; Error correction; Monitoring (error detection, correction or monitoring in information storage based on relative movement between record carrier and transducer G11B20/18; monitoring, i.e. supervising the progress of recording or reproducing G11B27/36; in static stores G11C29/00) · CPC title

• H04L41/147 Primary

  for predicting network behaviour · CPC title