Computation of optimal inventory targets in bill-of-materials in a multi-echelon supply chain
A system and method is disclosed for computing optimal inventory targets in bill of materials in a multi-echelon supply chain. The system includes one or more memory units and on or more processing units collectively configured to determine a delay time from at least one second plan level node in an at least second indenture plan level to at least one first plan level node in a first indenture plan level. The one or more memory units and on or more processing units is further collectively configured to determine a delay time from the at least one first plan level node in the first indenture plan level to a finished good.
1. A system to determine inventory targets in bill of materials in a multi-echelon supply chain network, comprising:
a supply chain network comprising two or more supply chain entities, two or more nodes, and two or more computers, the two or more nodes comprising an inventory of an item and the two or more computers each comprising one or more memory units and one or more processing units configured to:
determine a first delay time between at least two source nodes of a second indenture level and a destination node of a first indenture level;
determine a second delay time between the destination node of the first indenture level and a finished good, wherein the second delay time is based, at least in part, on increasing a mean and standard deviation of a lead time between the destination node of the first indenture level and the finished good by a mean and a standard deviation of a lead time of a bottleneck between the at least two source nodes of the second indenture level and the destination node of the first indenture level; and
determine an optimal inventory target based, at least in part, on the first and second delay times thereby reducing computational run time of at least one computer of the two or more computers, wherein at least one of the two or more supply chain entities adjusts the inventory based, at least in part, on the optimal inventory target.
2. The system of claim 1 , wherein the one or more processing units are further configured to:
determine the mean and the standard deviation of the lead time of the bottleneck by identifying the mean and the standard deviation of the lead time between the at least two source nodes of the second indenture level and the destination node of the first indenture level having a largest sum of a mean lead time and a multiplicative product of a lead time standard deviation and a sigma limit; and
select the mean and the standard deviation of the lead time between the at least two source nodes of the second indenture level and the destination node of the first indenture level having the largest sum as the mean and the standard deviation of the lead time of the bottleneck.
3. The system of claim 2 , wherein the one or more processing units are further configured to:
determine a coefficient of variation based, at least in part, on the mean and the standard deviation of the lead time between the at least two source nodes of the second indenture level and the destination node of the first indenture level, wherein the sigma limit is the coefficient of variation when the coefficient of variation is two or greater, and the sigma limit is two when the coefficient of variation is less than two.
4. The system of claim 2 , wherein the one or more processing units are further configured to:
determine a procurement problem variable standard deviation and a procurement problem variable based, at least in part, on the mean and standard deviation of the lead time, a mean daily demand, and a mean square error between the at least two source nodes of the second indenture level and the destination node of the first indenture level.
5. The system of claim 4 , wherein the one or more processing units are further configured to:
determine a coefficient of variation based, at least in part, on the procurement problem variable standard deviation divided by the procurement problem variable between the at least two source nodes of the second indenture level and the destination node of the first indenture level, wherein the sigma limit is the coefficient of variation when the coefficient of variation is two or greater, and the sigma limit is two when the coefficient of variation is less than two.
6. The system of claim 5 , wherein the at least two source nodes of the second indenture level are nodes of an at least second plan level and the destination node of the first indenture level is a node of an at least first plan level.
7. The system of claim 6 , wherein the one or more processing units are further configured to:
determine the mean daily demand of the destination node of the first indenture level based, at least in part, on a daily forecast for an independent demand of the finished good and any independent demand of a previous plan level.
8. A computer-implemented method of determining inventory targets in bill of materials in a multi-echelon supply chain network thereby reducing computational run time of a computer, the method comprising:
determining a first delay time between at least two source nodes of a second indenture level and a destination node of a first indenture level;
determining a second delay time between the destination node of the first indenture level and a finished good, wherein the second delay time is based, at least in part, on increasing a mean and a standard deviation of a lead time between the destination node of the first indenture level and the finished good by a mean and a standard deviation of a lead time of a bottleneck between the at least two source nodes of the second indenture level and the destination node of the first indenture level;
determining an optimal inventory target based, at least in part, on the first and second delay times thereby reducing computational run time of the computer; and
adjusting an inventory of at least one node based, at least in part, on the optimal inventory target.
9. The computer-implemented method of claim 8 , further comprising:
determining the mean and the standard deviation of the lead time of the bottleneck by identifying the mean and the standard deviation of the lead time between the at least two source nodes of the second indenture level and the destination node of the first indenture level having a largest sum of a mean lead time and a multiplicative product of a lead time standard deviation and a sigma limit; and
selecting the mean and the standard deviation of the lead time between the at least two source nodes of the second indenture level and the destination node of the first indenture level having the largest sum as the mean and the standard deviation of the lead time of the bottleneck.
10. The computer-implemented method of claim 9 , further comprising:
determining a coefficient of variation based, at least in part, on the mean and the standard deviation of the lead time between the at least two source nodes of the second indenture level and the destination node of the first indenture level, wherein the sigma limit is the coefficient of variation when the coefficient of variation is two or greater, and the sigma limit is two when the coefficient of variation is less than two.
11. The computer-implemented method of claim 9 , further comprising:
determining a procurement problem variable standard deviation and a procurement problem variable based, at least in part, on the mean and the standard deviation of the lead time, a mean daily demand, and a mean square error between the at least two source nodes of the second indenture level and the destination node of the first indenture level.
12. The computer-implemented method of claim 11 , further comprising:
determining a coefficient of variation based, at least in part, on the procurement problem variable standard deviation divided by the procurement problem variable between the at least two source nodes of the second indenture level and the destination node of the first indenture level, wherein the sigma limit is the coefficient of variation when the coefficient of variation is two or greater, and the sigma limit is two when the coefficient of variation is less than two.
13. The computer-implemented method of claim 12 , wherein the at least two source nodes of the second indenture level are nodes of an at least second plan level and the destination node of the first indenture level is a node of an at least first plan level.
14. The computer-implemented method of claim 13 , further comprising:
determining the mean daily demand of the destination node of the first indenture level based, at least in part, on a daily forecast for an independent demand of the finished good and any independent demand of a previous plan level.
15. A non-transitory computer-readable medium embodied with software to determine inventory targets in bill of materials in a multi-echelon supply chain network, the software when executed by one or more computers is configured to:
determine a first delay time between at least two source nodes of a second indenture level and a destination node of a first indenture level;
determine a second delay time between the destination node of the first indenture level and a finished good, wherein the second delay time is based, at least in part, on increasing a mean and a standard deviation of a lead time between the destination node of the first indenture level and the finished good by a mean and a standard deviation of a lead time of a bottleneck between the at least two source nodes of the second indenture level and the destination node of the first indenture level;
determine an optimal inventory target based, at least in part, on the first and second delay times thereby reducing computational run time of at least one computer of the one or more computers; and
adjust an inventory of at least one node based, at least in part, on the optimal inventory target.
16. The non-transitory computer-readable medium of claim 15 , wherein the software is further configured to:
determine the mean and the standard deviation of the lead time of the bottleneck by identifying the mean and the standard deviation of the lead time between the at least two source nodes of the second indenture level and the destination node of the first indenture level having a largest sum of a mean lead time and a multiplicative product of a lead time standard deviation and a sigma limit; and
select the mean and the standard deviation of the lead time between the at least two source nodes of the second indenture level and the destination node of the first indenture level having the largest sum as the mean and the standard deviation of the lead time of the bottleneck.
17. The non-transitory computer-readable medium of claim 16 , wherein the software is further configured to:
determine a coefficient of variation based, at least in part, on the mean and the standard deviation of the lead time between the at least two source nodes of the second indenture level and the destination node of the first indenture level, wherein the sigma limit is the coefficient of variation when the coefficient of variation is two or greater, and the sigma limit is two when the coefficient of variation is less than two.
18. The non-transitory computer-readable medium of claim 16 , wherein the software is further configured to:
determine a procurement problem variable standard deviation and a procurement problem variable based, at least in part, on the mean and the standard deviation of the lead time, a mean daily demand, and a mean square error between the at least two source nodes of the second indenture level and the destination node of the first indenture level.
19. The non-transitory computer-readable medium of claim 18 , wherein the software is further configured to:
determine a coefficient of variation based, at least in part, on the procurement problem variable standard deviation divided by the procurement problem variable between the at least two source nodes of the second indenture level and the destination node of the first indenture level, wherein the sigma limit is the coefficient of variation when the coefficient of variation is two or greater, and the sigma limit is two when the coefficient of variation is less than two.
20. The non-transitory computer-readable medium of claim 19 , wherein the at least two source nodes of the second indenture level are nodes of an at least second plan level and the destination node of the first indenture level is a node of an at least first plan level.