Optimizing an inventory of a supply chain

• US 20040230475A1
• Filed: 04/29/2004
• Published: 11/18/2004
• Est. Priority Date: 05/12/2003
• Status: Active Grant

• Current Assignee: Blue Yonder Group, Inc. (Panasonic Holdings Corporation)
• Original Assignee: i2 Technologies US Incorporated (Panasonic Holdings Corporation)
• Inventors: Najmi, Adeel, Dogan, Koray, Raman, Ramesh
• Granted Patent: US 7,853,462 B2
• Time in Patent Office: Days
• US Class Current: 705/10
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management
• CPC Class Codes: G06Q 10/06315 Needs-based resource requirements planning or analysis
• CPC Class Codes: G06Q 10/06375 Prediction of business process outcome or impact based on a proposed change
• CPC Class Codes: G06Q 10/087 Inventory or stock management, e.g. order filling, procurement or balancing against orders
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities

Abstract

Optimizing inventory targets for nodes of a supply chain to satisfy a target customer service level may include accessing a supply chain model that has an assumed value for each of a number of inputs. An optimized inventory target is calculated according to the supply chain model to satisfy the target customer service level, and a measured actual customer service level and a measured actual value for each input are accessed. If the measured actual customer service level fails to satisfy the target customer service level, deviations between the measured actual and assumed values for each input are determined. An input for which the deviation is significant is identified to be a root cause of the failure. For a subsequent time period, using the deviation for the identified input as feedback, the assumed value for the identified input is adjusted, and a reoptimized inventory target is calculated to satisfy the target customer service level.

Claims

1. A method for optimizing inventory targets for nodes in a multi-echelon supply chain to satisfy a target customer service level, comprising:

accessing a supply chain model comprising an assumed value for each of the following inputs for a first time period;

mean order lead time;

variability of order lead time;

for each of a number of order lead time intervals, mean demand within the order lead time interval;

for each of the number of order lead time intervals, variability of demand within the order lead time interval;

mean supply lead time for each of the nodes; and

variability of supply lead time for each of the nodes;

for the first time period, according to the supply chain model including the assumed values for the inputs, calculating an optimized inventory target for each of the nodes to satisfy the target customer service level;

for the first time period, accessing a measured actual customer service level and a measured actual value for each of the inputs;

if the measured actual customer service level for the first time period does not satisfy the target customer service level, then;

for each of the inputs, determining a deviation between the measured actual and assumed values for the input for the first time period; and

identifying at least one of the inputs for which the deviation for the first time period is significant to be a root cause of the measured actual customer service level for the first time period not satisfying the target customer service level; and

for a subsequent second time period, using the determined deviation for the identified input for the first time period as feedback;

adjusting the assumed value for the identified input in the supply chain model; and

calculating a reoptimized inventory target for each of the nodes to satisfy the target customer service level.

2. The method of claim 1, wherein the variability of order lead time, demand, and supply lead time comprises the standard deviation of order lead time, demand, and supply lead time, respectively.

3. The method of claim 1, wherein calculating the optimized inventory targets comprises:

propagating a portion of estimated demand on an end node in the supply chain to one or more upstream nodes in the supply chain according to estimated order lead time information;

calculating the optimized inventory target for each node according to the portion of the estimated demand on the node independently of the calculation of the optimized inventory target for every other node, the optimized inventory target for the end node being calculated according to the portion of the estimated demand not propagated to the one or more upstream nodes, the optimized inventory target for each of the one or more upstream nodes being calculated according to the portion of the estimated demand propagated to the upstream node.

4. The method of claim 3, wherein calculating the optimized inventory targets further comprises independently calculating the optimized inventory target for each of a plurality of nodes using a different computer processor in a distributed processing environment.

5. The method of claim 1, wherein the deviations between the measured actual and assumed values for the inputs are determined according to defined workflows that are consistent and repeatable over a plurality of successive time periods.

6. The method of claim 1, further comprising repeating the steps of calculating optimized inventory targets to satisfy the target customer service level, accessing a measured actual customer service level, if the measured actual customer service level does not satisfy the target customer service level then determining a deviation between the measured actual and assumed values for each input and identifying an input for which the deviation is significant as a root cause, adjusting the assumed value for the identified input, and calculating reoptimized inventory targets to satisfy the target customer service level in an iterative closed-loop process that is consistent and repeatable over a plurality of successive time periods, the iterative closed-loop process having the assumed values of the inputs as its inputs and the measured actual customer service level as its output.

7. A system for optimizing inventory targets for nodes in a multi-echelon supply chain to satisfy a target customer service level, comprising:

a database operable to store a supply chain model; and

a server system coupled to the database and operable to;

access the supply chain model comprising an assumed value for each of the following inputs for a first time period;

mean order lead time;

variability of order lead time;

for each of a number of order lead time intervals, mean demand within the order lead time interval;

for each of the number of order lead time intervals, variability of demand within the order lead time interval;

mean supply lead time for each of the nodes; and

variability of supply lead time for each of the nodes;

for the first time period, according to the supply chain model including the assumed values for the inputs, calculate an optimized inventory target for each of the nodes to satisfy the target customer service level;

for the first time period, access a measured actual customer service level and a measured actual value for each of the inputs;

if the measured actual customer service level for the first time period does not satisfy the target customer service level, then;

for each of the inputs, determine a deviation between the measured actual and assumed values for the input for the first time period; and

identify at least one of the inputs for which the deviation for the first time period is significant to be a root cause of the measured actual customer service level for the first time period not satisfying the target customer service level; and

for a subsequent second time period, using the determined deviation for the identified input for the first time period as feedback;

adjust the assumed value for the identified input in the supply chain model; and

calculate a reoptimized inventory target for each of the nodes to satisfy the target customer service level.

8. The system of claim 7, wherein the variability of order lead time, demand, and supply lead time comprises the standard deviation of order lead time, demand, and supply lead time, respectively.

9. The system of claim 7, wherein the server system is operable to calculate the optimized inventory targets by:

propagating a portion of estimated demand on an end node in the supply chain to one or more upstream nodes in the supply chain according to estimated order lead time information;

calculating the optimized inventory target for each node according to the portion of the estimated demand on the node independently of the calculation of the optimized inventory target for every other node, the optimized inventory target for the end node being calculated according to the portion of the estimated demand not propagated to the one or more upstream nodes, the optimized inventory target for each of the one or more upstream nodes being calculated according to the portion of the estimated demand propagated to the upstream node.

10. The system of claim 9, wherein the server system is operable to calculate the optimized inventory targets by independently calculating the optimized inventory target for each of a plurality of nodes using a different computer processor in a distributed processing environment.

11. The system of claim 7, wherein the deviations between the measured actual and assumed values for the inputs are determined according to defined workflows that are consistent and repeatable over a plurality of successive time periods.

12. The system of claim 7, the server system further operable to repeat the steps of calculating optimized inventory targets to satisfy the target customer service level, accessing a measured customer service level, if the measured actual customer service level does not satisfy the target customer service level then determining a deviation between the measured actual and assumed values for each input and identifying an input for which the deviation is significant as a root cause, adjusting the assumed value for the identified input, and calculating reoptimized inventory targets to satisfy the target customer service level in an iterative closed-loop process that is consistent and repeatable over a plurality of successive time periods, the iterative closed-loop process having the assumed values of the inputs as its inputs and the measured actual customer service level as its output.

13. Software for optimizing inventory targets for nodes in a multi-echelon supply chain to satisfy a target customer service level, the software embodied in a computer-readable medium and when executed by a computer operable to:

access a supply chain model comprising an assumed value for each of the following inputs for a first time period;

mean order lead time;

variability of order lead time;

for each of a number of order lead time intervals, mean demand within the order lead time interval;

for each of the number of order lead time intervals, variability of demand within the order lead time interval;

mean supply lead time for each of the nodes; and

variability of supply lead time for each of the nodes;

for the first time period, according to the supply chain model including the assumed values for the inputs, calculate an optimized inventory target for each of the nodes to satisfy the target customer service level;

for the first time period, access a measured actual customer service level and a measured actual value for each of the inputs;

if the measured actual customer service level for the first time period does not satisfy the target customer service level, then;

for each of the inputs, determine a deviation between the measured actual and assumed values for the input for the first time period; and

identify at least one of the inputs for which the deviation for the first time period is significant to be a root cause of the measured actual customer service level for the first time period not satisfying the target customer service level; and

for a subsequent second time period, using the determined deviation for the identified input for the first time period as feedback;

adjust the assumed value for the identified input in the supply chain model; and

calculate a reoptimized inventory target for each of the nodes to satisfy the target customer service level.

14. The software of claim 13, wherein the variability of order lead time, demand, and supply lead time comprises the standard deviation of order lead time, demand, and supply lead time, respectively.

15. The software of claim 13, operable to calculate the optimized inventory targets by:

propagating a portion of estimated demand on an end node in the supply chain to one or more upstream nodes in the supply chain according to estimated order lead time information;

calculating the optimized inventory target for each node according to the portion of the estimated demand on the node independently of the calculation of the optimized inventory target for every other node, the optimized inventory target for the end node being calculated according to the portion of the estimated demand not propagated to the one or more upstream nodes, the optimized inventory target for each of the one or more upstream nodes being calculated according to the portion of the estimated demand propagated to the upstream node.

16. The software of claim 15, operable to calculate the optimized inventory targets by independently calculating the optimized inventory target for each of a plurality of nodes using a different computer processor in a distributed processing environment.

17. The software of claim 13, wherein the deviations between the measured actual and assumed values for the inputs are determined according to defined workflows that are consistent and repeatable over a plurality of successive time periods.

18. The software of claim 13, further operable to repeat the steps of calculating optimized inventory targets to satisfy the target customer service level, accessing a measured actual customer service level, if the measured actual customer service level does not satisfy the target customer service level then determining a deviation between the measured actual and assumed values for each input and identifying an input for which the deviation is significant as a root cause, adjusting the assumed value for the identified input, and calculating reoptimized inventory targets to satisfy the target customer service level in an iterative closed-loop process that is consistent and repeatable over a plurality of successive time periods, the iterative closed-loop process having the assumed values of the inputs as its inputs and the measured actual customer service level as its output.

19. A system for optimizing inventory targets for nodes in a multi-echelon supply chain to satisfy a target customer service level, comprising:

means for accessing a supply chain model comprising an assumed value for each of the following inputs for a first time period;

mean order lead time;

variability of order lead time;

for each of a number of order lead time intervals, mean demand within the order lead time interval;

for each of the number of order lead time intervals, variability of demand within the order lead time interval;

mean supply lead time for each of the nodes; and

variability of supply lead time for each of the nodes;

for the first time period, according to the supply chain model including the assumed values for the inputs, means for calculating an optimized inventory target for each of the nodes to satisfy the target customer service level;

for the first time period, means for accessing a measured actual customer service level and a measured actual value for each of the inputs;

if the measured actual customer service level for the first time period does not satisfy the target customer service level, then;

for each of the inputs, means for determining a deviation between the measured actual and assumed values for the input for the first time period; and

means for identifying at least one of the inputs for which the deviation for the first time period is significant to be a root cause of the measured actual customer service level for the first time period not satisfying the target customer service level; and

for a subsequent second time period, using the determined deviation for the identified input for the first time period as feedback;

means for adjusting the assumed value for the identified input in the supply chain model; and

means for calculating a reoptimized inventory target for each of the nodes to satisfy the target customer service level.

20. A method for optimizing inventory targets for nodes in a multi-echelon supply chain to satisfy a target customer service level, comprising:

accessing a supply chain model comprising an assumed value for each of the following inputs for a first time period;

mean order lead time;

variability of order lead time;

for each of a number of order lead time intervals, mean demand within the order lead time interval;

for each of the number of order lead time intervals, variability of demand within the order lead time interval;

mean supply lead time for each of the nodes; and

variability of supply lead time for each of the nodes, the variability of order lead time, demand, and supply lead time comprising the standard deviation of order lead time, demand, and supply lead time, respectively;

for the first time period, according to the supply chain model including the assumed values for the inputs, calculating an optimized inventory target for each of the nodes to satisfy the target customer service level by;

propagating a portion of estimated demand on an end node in the supply chain to one or more upstream nodes in the supply chain according to estimated order lead time information;

calculating the optimized inventory target for each node according to the portion of the estimated demand on the node independently of the calculation of the optimized inventory target for every other node, the optimized inventory target for the end node being calculated according to the portion of the estimated demand not propagated to the one or more upstream nodes, the optimized inventory target for each of the one or more upstream nodes being calculated according to the portion of the estimated demand propagated to the upstream node, the optimized inventory target for each of a plurality of nodes independently calculated using a different computer processor in a distributed processing environment;

for the first time period, accessing a measured actual customer service level and a measured actual value for each of the inputs;

if the measured actual customer service level for the first time period does not satisfy the target customer service level, then;

for each of the inputs, determining a deviation between the measured actual and assumed values for the input for the first time period, the deviation between the measured actual and assumed values for the inputs determined according to defined workflows that are consistent and repeatable over a plurality of successive time periods; and

identifying at least one of the inputs for which the deviation for the first time period is significant to be a root cause of the measured actual customer service level for the first time period not satisfying the target customer service level;

for a subsequent second time period, using the determined deviation for the identified input for the first time period as feedback;

adjusting the assumed value for the identified input in the supply chain model; and

calculating a reoptimized inventory target for each of the nodes to satisfy the target customer service level; and

repeating the steps of calculating optimized inventory targets to satisfy the target customer service level, accessing a measured actual customer service level, if the measured actual customer service level does not satisfy the target customer service level then determining a deviation between the measured actual and assumed values for each input and identifying an input for which the deviation is significant as a root cause, adjusting the assumed value for the identified input, and calculating reoptimized inventory targets to satisfy the target customer service level in an iterative closed-loop process that is consistent and repeatable over a plurality of successive time periods, the iterative closed-loop process having the assumed values of the inputs as its inputs and the measured actual customer service level as its output.