Managing supply chains with model predictive control

• US 20040267394A1
• Filed: 06/30/2003
• Published: 12/30/2004
• Est. Priority Date: 06/30/2003
• Status: Active Grant

• Current Assignee: Intel Corporation, Arizona Board of Regents (University of Arizona)
• Original Assignee: Intel Corporation
• Inventors: Rivera, Daniel E., Smith, Kirk, Wang, Wenlin, Braun, Martin W., Kempf, Karl, Toperzer, Beverly
• Granted Patent: US 7,054,706 B2
• Time in Patent Office: Days
• US Class Current: 700/99
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders

Abstract

Entering a demand prediction for a discrete component product manufactured in a supply chain into a model predictive controller, entering inventory information for an inventory in the supply chain that is associated with the discrete component product into the model predictive controller, and controlling the inventory with the model predictive controller based on the demand prediction is disclosed.

Claims

1. A method comprising:

entering a demand prediction for a discrete component product manufactured in a supply chain into a model predictive controller;

entering inventory information for an inventory in the supply chain that is associated with the discrete component product into the model predictive controller; and

controlling the inventory with the model predictive controller based on the demand prediction.

2. The method of claim 1, further comprising:

predicting the demand for the discrete component product.

3. The method of claim 2, wherein the predicted demand is for a future time that is related to a throughput time of the supply chain.

4. The method of claim 3, wherein the future time is proportional to the throughput time.

5. The method of claim 1, further comprising:

partitioning an intermediate inventory into a plurality of smaller inventories that each correspond to different discrete component products with the model predictive controller based on the demand prediction, wherein the plurality of smaller inventories include the controlled inventory, and wherein the controlled inventory corresponds to a product associated with the demand prediction.

6. The method of claim 1, further comprising:

partitioning an intermediate inventory of microelectronic devices into the controlled inventory and a second inventory with the model predictive controller based on the demand prediction, wherein the controlled inventory corresponds to a first type of microelectronic package and the second inventory corresponds to a second type of microelectronic package, and wherein the demand prediction corresponds to the first type of microelectronic package.

7. The method of claim 1, further comprising:

controlling a plurality of inventories of different discrete components that are associated with products that each have related demands with the discrete component product with the model predictive controller based on the predicted demand for the discrete component product.

8. The method of claim 7, wherein the plurality of inventories reside at different manufacturing facilities of the supply chain.

9. The method of claim 1, wherein controlling the inventory includes controlling an inventory of discrete components associated with processors with the model predictive controller based on the demand prediction, and wherein the demand prediction includes a demand prediction for the processors.

10. The method of claim 9, further comprising:

controlling an inventory of chipset microelectronic devices having related demand with the processors with the model predictive controller based on the demand prediction for the processors.

11. A machine-readable medium having stored thereon data representing sequences of instructions that, if executed, cause a machine to:

receive a demand prediction for a discrete component product manufactured in a supply chain;

receive inventory information for an inventory in the supply chain that is associated with the discrete component product; and

control the inventory based on model predictive control, and based on the demand prediction.

12. The machine-readable medium of claim 11, wherein the instructions further comprise instructions that, if executed, cause the machine to:

predict the demand for the discrete component product at a future time that is related to a throughput time of the supply chain.

13. The machine-readable medium of claim 11, wherein the instructions further comprise instructions that, if executed, cause the machine to:

partition an intermediate inventory into a plurality of smaller inventories that each correspond to different discrete component products based on model predictive control and based on the demand prediction, wherein the plurality of smaller inventories include the controlled inventory, and wherein the controlled inventory corresponds to a product associated with the demand prediction.

14. The machine-readable medium of claim 11, wherein the instructions further comprise instructions that, if executed, cause the machine to:

partition an intermediate inventory of microelectronic devices into the controlled inventory and a second inventory based on model predictive control and based on the demand prediction, wherein the controlled inventory corresponds to a first type of microelectronic package and the second inventory corresponds to a second type of microelectronic package, and wherein the demand prediction corresponds to the first type of microelectronic package.

15. The machine-readable medium of claim 11, wherein the instructions further comprise instructions that, if executed, cause the machine to:

control a plurality of inventories of different discrete components that are associated with products that each have related demands with the discrete component product based on model predictive control and based on the predicted demand for the discrete component product.

16. An apparatus comprising:

a memory to store inventory information for a discrete component inventory, inventory information for a discrete component product inventory, and a demand prediction for a discrete component product; and

a model predictive controller to control the discrete component inventory based on model predictive control, based on the inventory information for the inventories, and based on the predicted demand for the discrete component.

17. The apparatus of claim 16, further comprising:

an inventory monitor to monitor and report inventory information for the discrete component inventory to the model predictive controller.

18. The apparatus of claim 16, wherein the model predictive controller further comprises:

an inventory partitioner to partition a second discrete component inventory up process from the controlled discrete component inventory into a plurality of smaller inventories that each correspond to different discrete component products based on model predictive control, and based on the demand prediction, wherein the plurality of smaller inventories include the controlled discrete component inventory.

19. The apparatus of claim 16, wherein the model predictive controller further comprises:

a multiple inventory controller to control a plurality of inventories of different discrete components that are associated with products that each have related demands with the discrete component product with the model predictive controller based on the predicted demand for the discrete component product.

20. A supply chain management system comprising:

an inventory system to maintain inventory information for a discrete component inventory; and

an apparatus comprising;

a communication device to provide a communication link to a network to receive the inventory information;

a memory to store the received inventory information and a demand prediction for a discrete component product associated with the discrete component inventory; and

a model predictive controller to control the discrete component inventory based on model predictive control, based on the inventory information, and based on the predicted demand for the discrete component product.

21. The system of claim 20, further comprising:

a demand predictor to predict a demand for the discrete component product at a future time that is related to a supply chain throughput time.

22. The system of claim 20, wherein the model predictive controller further comprises:

an inventory partitioner to partition a second discrete component inventory up process from the controlled discrete component inventory into a plurality of smaller inventories that each correspond to different discrete component products based on model predictive control, and based on the demand prediction, wherein the plurality of smaller inventories include the controlled discrete component inventory.

23. The system of claim 20, wherein the model predictive controller further comprises:

a multiple inventory controller to control a plurality of inventories of different discrete components that are associated with products that each have related demands with the discrete component product with the model predictive controller based on the predicted demand for the discrete component product.