System and method for controlling the number of units of parts in an inventory

• US 5,608,621 A
• Filed: 03/24/1995
• Issued: 03/04/1997
• Est. Priority Date: 03/24/1995
• Status: Expired due to Term

• Current Assignee: Panduit Corporation
• Original Assignee: Panduit Corporation
• Inventors: Caveney, Jack, Winger, James
• Primary Examiner(s): McElheny, Jr., Donald E.
• Application Number:
  US08/410,342
• Time in Patent Office: 711 Days
• Field of Search: 364/401 R, 364/400, 364/468.01, 364/468.05, 364/468.09, 364/478.04, 364/478.02
• US Class Current: 705/7.25
• 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

A computer system according to the present invention comprises in a preferred embodiment a memory, a processor, an input and an output. The computer system receives part data and forecast data for each part in an inventory, including the number of units in the inventory, the cost, and the forecast unit demand. The computer system uses the part data and forecast data for each part to optimize the number of units of each part in the inventory for a selected inventory investment or service level constraint. The computer system optimizes the number of units of each part in the inventory by determining the number of units for each part for which the marginal increase in the number of orders for the part filled from the inventory for each additional order stocked is equal to the marginal increase for every other part and for which either the inventory service level equals the service level constraint or the inventory investment equals the inventory investment constraint. The computer system then outputs a determined minimum unit replenishment quantity and a safety unit quantity for each part. The minimum unit replenishment quantities and the safety unit quantity are used to determine the number of units of each part used to resupply the inventory.

Claims

1. A computer system for automatically controlling the number of units of each of a plurality of different parts in an inventory having a predetermined inventory investment constraint, the computer system comprising:

means for receiving part data for each part in the inventory, the part data for each part including the number of units of the part in the inventory, the cost of the part, and a historical average ratio of units per order for the part, each order being for at least one unit of the part;

means for storing the part data for each part;

means for inputting a plurality of part service levels for each part, each part service level for each part representing an expected fraction of the orders for the part which can be filled from the inventory;

means for retrieving the part data for each part and for determining a part data table for each part which relates forecast data for the part to the part data for the part and the plurality of part service levels for the part, the forecast data for each part including a forecast unit demand for the part and an average number of units of the part for each of the plurality of part service levels for the part;

means for storing the part data table for each part;

means for retrieving the part data table for each part and for determining a slope table for each part which relates each of the plurality of part service levels for the part to an expected number of fillable-from-stock orders for the part and a slope for the part, each expected number of fillable-from-stock orders for the part being the product of the corresponding part service level and the ratio of the forecast unit demand for the part to the historical average ratio of units per order for the part, each slope for the part being the ratio of the change in the corresponding expected number of fillable-from-stock orders for the part to the change in the ratio of the corresponding average number of units of the part expected to be in the inventory to the historical average ratio of units per order for the part;

means for storing the slope table for each part;

means for retrieving the slope tables for all the parts and for determining a slope which is common to each slope table and for which the sum of expected part investments for each part corresponding to the slope is equal to the inventory investment constraint, each expected part investment being the product of the corresponding average number of units of the part expected to be in the inventory and the cost of the part;

means for determining a minimum unit replenishment quantity and a safety unit quantity for each part capable of effecting the part service level for the part corresponding to the determined common slope;

means for outputting the minimum unit replenishment quantity and the safety unit quantity for each part; and

means for ordering the minimum unit replenishment quantity for at least one of the parts from a supply source when the number of units of the part in the inventory is less than or equal to the safety unit quantity.

2. A computer system for controlling the number of units of each of a plurality of different parts in an inventory, the computer system comprising:

a memory configured to store a part data table for each part, wherein the part data table for each part relates part data for the part to forecast data for the part, the part data for each part including the number of units of the part in the inventory, the cost of the part, and a historical average ratio of units per order for the part, the forecast data for each part including a forecast unit demand for the part and an average number of units of the part expected to be in the inventory for each of a plurality of part service levels for the part, each part service level for each part representing an expected fraction of the orders for the part which can be filled from the inventory;

an input device for receiving a selected inventory investment constraint;

a processor operatively coupled to the memory and the input device, the processor configured to;

retrieve the part data table for each part,

determine a slope table for each part which relates each of the plurality of part service levels for the part to an expected number of fillable-from-stock orders for the part and a slope for the part, each expected number of fillable-from-stock orders for the part being the product of the corresponding part service level and the ratio of the forecast unit demand for the part to the historical average ratio of units per order for the part, each slope for the part being the ratio of the change in the corresponding expected number of fillable-from-stock orders for the part to the change in the ratio of the corresponding average number of units of the part expected to be in the inventory to the historical average ratio of units per order for the part,

store the slope table for each part in the memory,

retrieve the slope tables for all the parts,

determine a slope which is common to each slope table and for which the sum of expected part investments for each part corresponding to the slope is equal to the selected inventory investment constraint, each expected part investment being the product of the corresponding average number of units of the part expected to be in the inventory and the cost of the part, and

determine a minimum unit replenishment quantity and a safety unit quantity for each part capable of effecting the part service level for the part corresponding to the determined common slope; and

an output device operatively coupled to the memory and the processor, the output device configured for outputting the minimum unit replenishment quantity and the safety unit quantity for each part.

3. The computer system of claim 2 wherein the forecast unit demand for each part is normally distributed.

4. A computer system for controlling the number of units of each of a plurality of different parts in an inventory, the computer system comprising:

a memory configured to store a part data table for each part, wherein the part data table for each part relates part data for the part to forecast data for the part, the part data for each part including the number of units of the part in the inventory and a historical average ratio of units per order for the part, each order being for at least one unit of the part, the forecast data for each part including a forecast unit demand for the part and an average number of units of the part expected to be in the inventory for each of a plurality of part service levels for the part, each part service level for each part representing an expected fraction of the orders for the part which can be filled from the inventory;

an input device for receiving a selected inventory service level constraint;

a processor operatively coupled to the memory and the input device, the processor configured to;

retrieve the part data table for each part,

determine a slope table for each part which relates each of the plurality of part service levels for the part to an expected number of orders for the part, an expected number of fillable-from-stock orders for the part, and a slope for the part, each expected number of orders for the part being the ratio of the forecast unit demand to the historical average ratio of units per order for the part, each expected number of fillable-from-stock orders for the part being the product of the corresponding part service level and the corresponding expected number of orders for the part, each slope for the part being the ratio of the change in the corresponding expected number of fillable-from-stock orders for the part to the change in the ratio of the corresponding average number of units of the part expected to be in the inventory to the historical average ratio of units per order for the part,

store the slope table for each part in the memory,

retrieve the slope tables for all the parts,

determine a slope which is common to each slope table and for which the ratio of the sum of the corresponding expected number of fillable-from-stock orders for all the parts to the sum of the expected number of orders for all the parts is equal to the selected inventory service level constraint, and

determine a minimum unit replenishment quantity and a safety unit quantity for each part capable of effecting the part service level for the part corresponding to the determined common slope; and

an output device operatively coupled to the memory and the processor, the output device configured for outputting the minimum unit replenishment quantity and the safety unit quantity for each part.

5. The computer system of claim 4 wherein the forecast unit demand for each part is normally distributed.

6. A computer system for controlling the number of units of each of a plurality of different parts in an inventory, the computer system comprising:

a memory configured to store a part data table for each part, wherein the part data table for each part relates part data for the part to forecast data for the part, the part data for each part including the number of units of the part in the inventory, the cost of the part, and a historical average ratio of units per order for the part, each order being for at least one unit of the part, the forecast data for each part including a forecast unit demand for the part and an average number of units of the part expected to be in the inventory for each of a plurality of part service levels for the part, each part service level for each part representing an expected fraction of the orders for the part which can be filled from the inventory;

an input device for receiving a selected inventory investment constraint;

a processor operatively coupled to the memory and the input device, the processor configured to;

retrieve the part data table for each part,

determine a slope table for each part which relates each of the plurality of part service levels for the part to an expected number of fillable-from-stock orders for the part and a slope for the part, each expected number of fillable-from-stock orders for the part being the product of the corresponding part service level and the ratio of the forecast unit demand for the part to the historical average ratio of units per order for the part, each slope for the part being the ratio of the corresponding expected number of fillable-from-stock orders for the part to the ratio of the corresponding average number of units of the part expected to be in the inventory to the historical average ratio of units per order for the part,

store the slope table for each part in the memory,

retrieve the slope tables for all the parts,

determine a slope which is common to each slope table and for which the sum of expected part investments for each part corresponding to the slope is equal to the selected inventory investment constraint, each expected part investment being the product of the corresponding average number of units of the part expected to be in the inventory and the cost of the part, and

determine a minimum unit replenishment quantity and a safety unit quantity for each part capable of effecting the part service level for the part corresponding to the determined common slope; and

an output device operatively coupled to the memory and the processor, the output device configured for outputting the minimum unit replenishment quantity and the safety unit quantity for each part.

7. A method implemented on a computer system for controlling the number of units of each of a plurality of different parts in a vendor-managed inventory, the computer system comprising a memory, an input device, an output device, and a processor operatively coupled to the memory, the input device, and the output device, the method comprising:

collecting part data for each part in the inventory, the part data for each part including the number of units of the part in the inventory, the cost of the part, and a historical average ratio of units per order for the part, each order being for at least one unit of the part;

receiving and storing the part data for each part;

receiving a plurality of part service levels for each part and a selected inventory investment constraint, each part service level for each part representing an expected fraction of the orders for the part which can be filled from the inventory;

retrieving the part data for each part and determining a part data table for each part which relates forecast data for the part to the part data for the part and the plurality of part service levels for the part, the forecast data for each part including a forecast unit demand for the part and an average number of units of the part expected to be in the inventory for each of the plurality of part service levels for the part;

storing the part data table for each part;

retrieving the part data table for each part and determining a slope table for each part which relates each of the plurality of part service levels for the part to an expected number of fillable-from-stock orders for the part and a slope for the part, each expected number of fillable-from-stock orders for the part being the product of the corresponding part service level and the ratio of the forecast unit demand for the part to the historical average ratio of units per order for the part, each slope for the part being the ratio of the change in the corresponding expected number of fillable-from-stock orders for the part to the change in the ratio of the corresponding average number of units of the part expected to be in the inventory to the historical average ratio of units per order for the part;

storing the slope table for each part;

retrieving the slope tables for all the parts and determining a slope which is common to each slope table and for which the sum of expected part investments for each part corresponding to the slope is equal to the selected inventory investment constraint, each expected part investment being the product of the corresponding average number of units of the part expected to be in the inventory and the cost of the part;

determining a minimum unit replenishment quantity and a safety unit quantity for each part capable of effecting the part service level for the part corresponding to the determined common slope;

outputting the minimum unit replenishment quantity and the safety unit quantity for each part; and

when the number of units of a part in the vendor-managed inventory is less than or equal to the safety unit quantity, supplying at least a number of units of the part equal to the minimum unit replenishment quantity for the part from a supply source to the vendor-managed inventory.

8. The method of claim 7 wherein the forecast unit demand for each part is normally distributed.

9. A method implemented on a computer system for controlling the number of units of each of a plurality of different parts in an inventory, the computer system comprising a memory, an input device, an output device, and a processor operatively coupled to the memory, the input device, and the output device, the method comprising:

receiving a selected inventory service level constraint;

receiving and storing a part data table for each part, wherein each part data table relates part data for the part to forecast data for the part, the part data for each part including the number of units of the part in the inventory and a historical average ratio of units per order for the part, each order being for at least one unit of the part, the forecast data for each part including a forecast unit demand for the part and an average number of units of the part expected to be in the inventory for each of a plurality of part service levels for the part, each part service level for each part representing an expected fraction of the orders for the part which can be filled from the inventory;

retrieving the part data table for each part and determining a slope table for each part which relates each of the plurality of part service levels for the part to an expected number of orders for the part, an expected number of fillable-from-stock orders for the part, and a slope for the part, each expected number of orders for the part being the ratio of the forecast unit demand for the part to the historical average ratio of units per order for the part, each expected number of fillable-from-stock orders for the part being the product of the corresponding part service level and the expected number of orders for the part, each slope for the part being the ratio of the change in the corresponding expected number of fillable-from-stock orders for the part to the change in the ratio of the corresponding average number of units of the part expected to be in the inventory to the historical average ratio of units per order for the part;

storing the slope table for each part;

retrieving the slope tables for all the parts and determining a slope which is common to each slope table and for which the ratio of the sum of the corresponding expected number of fillable-from-stock orders for all the parts to the sum of the expected number of orders for all the parts is equal to the selected inventory service level constraint;

determining a minimum unit replenishment quantity and a safety unit quantity for each part capable of effecting the part service level for the part corresponding to the determined common slope; and

outputting the minimum unit replenishment quantity and the safety unit quantity for each part.

10. The method of claim 9 wherein the forecast unit demand for each part is normally distributed.

11. A method implemented on a computer system for controlling the number of units of each of a plurality of different parts in an inventory, the computer system comprising a memory, an input device, an output device, and a processor operatively coupled to the memory, the input device, and the output device, the method comprising:

receiving a selected inventory investment constraint;

receiving and storing a part data table for each part, wherein each part data table relates part data for the part to forecast data for the part, the part data for each part including the number of units of the part in the inventory, the cost of the part, and a historical average ratio of units per order for the part, each order being for at least one unit of the part, the forecast data for each part including a forecast unit demand for the part and an average number of units of the part expected to be in the inventory for each of a plurality of part service levels for the part, each part service level for each part representing an expected fraction of the orders for the part which can be filled from the inventory;

retrieving the part data table for each part and determining a slope table for each part which relates each of the plurality of part service levels for the part to an expected number of fillable-from-stock orders for the part and a slope for the part, each expected number of fillable-from-stock orders for the part being the product of the corresponding part service level and the ratio of the forecast unit demand for the part to the historical average ratio of units per order for the part, each slope for the part being the ratio of the corresponding expected number of fillable-from-stock orders for the part to the ratio of the corresponding average number of units of the part expected to be in the inventory to the historical average ratio of units per order for the part;

storing the slope table for each part;

retrieving the slope tables for all the parts and determining a slope which is common to each slope table and for which the sum of expected part investments for each part corresponding to the slope is equal to the selected inventory investment constraint, each expected part investment being the product of the corresponding average number of units of the part expected to be in the inventory and the cost of the part;

determining a minimum unit replenishment quantity and a safety unit quantity for each part capable of effecting the part service level for the part corresponding to the determined common slope; and

outputting the minimum unit replenishment quantity and the safety unit quantity for each part.

12. A method for controlling the number of units of each of a plurality of different parts in an inventory, the method comprising operating a computer to receive a selected inventory investment constraint, further operating the computer to receive and store a part data table for each part, wherein each part data table relates part data for the part to forecast data for the part, the part data for each part including the number of units of the part in the inventory, the cost of the part, and a historical average ratio of units per order for the part, each order being for at least one unit of the part, the forecast data including a forecast unit demand for the part and an average number of units of the part expected to be in the inventory for each of a plurality of part service levels for the part, each part service level for each part representing an expected fraction of the orders for the part which can be filled from the inventory, further operating the computer to retrieve the part data table for each part and determine a slope table for each part which relates each of the plurality of part service levels for the part to an expected number of fillable-from-stock orders for the part and a slope for the part, each expected number of fillable-from-stock orders for the part being the product of the corresponding part service level and the ratio of the forecast unit demand for the part to the historical average ratio of units per order for the part, each slope for the part being the ratio of the change in the corresponding expected number of fillable-from-stock orders for the part to the change in the ratio of the corresponding average number of units of the part expected to be in the inventory to the historical average ratio of units per order for the part, further operating the computer to store the slope table for each part, further operating the computer to retrieve the slope tables for all the parts and determine a slope which is common to each slope table and for which the sum of expected part investments for each part corresponding to the slope is equal to the selected inventory investment constraint, each expected part investment being the product of the corresponding average number of units of the part expected to be in the inventory and the cost of the part, further operating the computer to determine a minimum unit replenishment quantity and a safety unit quantity for each part capable of effecting the part service level for the part corresponding to the determined common slope, and further operating the computer to output the minimum unit replenishment quantity and the safety unit quantity for each part.

13. The method of claim 12 wherein the forecast unit demand for each part is normally distributed.

14. A method implemented on a computer system for controlling the number of units of each of a plurality of different parts in an inventory having an inventory investment constraint, wherein the computer system comprises a memory, a processor operatively coupled to the memory, and an output operatively coupled to the memory and the processor, the method comprising the steps of:

providing a part data table for each part stored in the memory;

retrieving the part data table for each part from the memory with the processor;

determining for each part with the processor a plurality of average numbers of units of the part expected to be in the inventory and a plurality of expected numbers of fillable-from-stock orders for the part corresponding to the part data table for the part;

determining for all the parts with the processor the average number of units of the parts expected to be in the inventory for which corresponding marginal changes in the expected number of fillable-from-stock orders for each of the parts are equal and for which the sum of expected part investments for each of the parts is equal to the inventory investment constraint;

determining for each part with the processor a minimum unit replenishment quantity and a safety unit quantity capable of effecting the determined average number of units of the part expected to be in the inventory; and

outputting on the output the minimum unit replenishment quantity and the safety unit quantity for each part.

15. The method of claim 14 wherein the marginal change in the expected number of fillable-from-stock orders for each part is a weighted average of at least a first factor and a second factor, wherein the first factor is the change expected in the expected number of fillable-from-stock orders for the part for each change in the product of the average number of units of the part expected to be in the inventory times the cost of the part, wherein the second factor is the change expected in the expected number of fillable-from-stock orders for the part for each change in the ratio of the average number of units of the part expected to be in the inventory to a historical ratio of units per order for the part.

16. The method of claim 15 wherein the first and second factors are normalized.

17. The method of claim 14 wherein the step of determining the average number of units of the parts expected to be in the inventory comprises:

determining a plurality of slopes for each part which are equal to the ratio of the rate of change of the expected number of fillable-from-stock orders for the part to the rate of change of the ratio of the corresponding average number of units of the part expected to be in the inventory to a historical average ratio of units per order for the part;

determining a slope from the plurality of slopes for each part which is common to each part and for which the sum of the expected part investments for each of the parts is equal to the inventory investment constraint; and

determining the average number of units of each part expected to be in the inventory corresponding to the determined common slope.

18. A method implemented on a computer system for controlling the number of units of each of a plurality of different parts in an inventory, the computer system comprising a memory, an input device, an output device, and a processor operatively coupled to the memory, the input device, and the output device, the method comprising:

collecting part data for each part in the inventory, the part data for each part including the number of units of the part in the inventory, the cost of the part, and a historical average ratio of units per order for the part, each order being for at least one unit of the part;

receiving and storing the part data for each part;

receiving a plurality of part service levels for each part and a selected inventory investment constraint, each part service level for each part representing an expected fraction of the orders for the part which can be filled from the inventory;

retrieving the part data for each part and determining a part data table for each part which relates forecast data for the part to the part data for the part and the plurality of part service levels for the part, the forecast data for each part including a forecast unit demand for the part and an average number of units of the part expected to be in the inventory for each of the plurality of part service levels for the part;

storing the part data table for each part;

retrieving the part data table for each part and determining a slope table for each part which relates each of the plurality of part service levels for the part to an expected number of fillable-from-stock orders for the part and a slope for the part, each expected number of fillable-from-stock orders for the part being the product of the corresponding part service level and the ratio of the forecast unit demand for the part to the historical average ratio of units per order for the part, each slope for the part being the ratio of the change in the corresponding expected number of fillable-from-stock orders for the part to the change in a corresponding expected part investment, each expected part investment being the product of the corresponding average number of units of the part expected to be in the inventory and the cost of the part;

storing the slope table for each part;

retrieving the slope tables for all the parts and determining a slope which is common to each slope table and for which the sum of the expected part investments for each part corresponding to the slope is equal to the selected inventory investment constraint;

determining a minimum unit replenishment quantity and a safety unit quantity for each part capable of effecting the part service level for the part corresponding to the determined common slope;

outputting the minimum unit replenishment quantity and the safety unit quantity for each part; and

when the number of units of a part in the inventory is less than or equal to the safety unit quantity, supplying at least a number of units of the part equal to the minimum unit replenishment quantity for the part from a supply source to the inventory.

19. The method of claim 18 wherein the forecast unit demand for each part is normally distributed.

20. A method implemented on a computer system for controlling the number of units of each of a plurality of different parts in an inventory, the computer system comprising a memory, an input device, an output device, and a processor operatively coupled to the memory, the input device, and the output device, the method comprising:

collecting part data for each part in the inventory, the part data for each part including the number of units of the part in the inventory and the cost of the part;

receiving and storing the part data for each part;

receiving a plurality of part service levels for each part and a selected inventory investment constraint, each part service level for each part representing an expected fraction of ordered units of the part which can be supplied from the inventory;

retrieving the part data for each part and determining a part data table for each part which relates forecast data for the part to the part data for the part and the plurality of part service levels for the part, the forecast data for each part including a forecast unit demand for the part and an average number of units of the part expected to be in the inventory for each of the plurality of part service levels for the part;

storing the part data table for each part;

retrieving the part data table for each part and determining a slope table for each part which relates each of the plurality of part service levels for the part to an expected number of fillable-from-stock ordered units for the part and a slope for the part, each expected number of fillable-from-stock ordered units for the part being the product of the corresponding part service level and the forecast unit demand for the part, each slope for the part being the ratio of the change in the corresponding expected number of fillable-from-stock ordered units for the part to the change in the corresponding average number of units of the part expected to be in the inventory;

storing the slope table for each part;

retrieving the slope tables for all the parts and determining a slope which is common to each slope table and for which the sum of expected part investments for each part corresponding to the slope are equal to the selected inventory investment constraint, each expected part investment being the product of the corresponding average number of units of the part expected to be in the inventory and the cost of the part;

determining a minimum unit replenishment quantity and a safety unit quantity for each part capable of effecting the part service level for the part corresponding to the determined common slope;

outputting the minimum unit replenishment quantity and the safety unit quantity for each part; and

when the number of units of a part in the inventory is less than or equal to the safety unit quantity, supplying at least a number of units of the part equal to the minimum unit replenishment quantity for the part from a supply source to the inventory.

21. The method of claim 20 wherein the forecast unit demand for each part is normally distributed.