Managing stock inventory levels
View Patent ↗The present invention provides a method of managing a stock replenishment system in a way that explicitly controls the in-stock rate. Also provided by the present invention is a method of managing a collection of stock replenishment systems in a way that the combined in-stock rate of the collection is controlled to a selected value. In addition, the present invention provides a method to set up a collection of stock replenishment systems in replenishment software systems in an identical fashion while controlling the combined in-stock rate. The methods of the present invention use point-of-sale data to determine packages-per-shopper rates, which are used as one of the inputs to statistical models of multiple-day retail sales.
1. A method comprising:
receiving a selection of an acceptable out-of-stock probability for an item, the acceptable out-of-stock probability indicating an acceptable risk of the item going out-of-stock at a particular retail store;
accessing sales data for the item, wherein the sales data is for a period of time;
determining, using a computing device, a quantity of packages of the item sold during the period of time based at least partly on the sales data;
determining a quantity of shoppers per day that purchased the item based at least partly on the sales data;
determining, using the computing device, a maximal sales estimate indicating a volume of sales that will cause an out-of-stock probability for the item to exceed the acceptable out-of-stock probability, wherein the maximal sales estimate is based on at least the determined quantity of packages of the item sold during the period of time, the determined shoppers per day, the acceptable out-of-stock probability, and a reorder time for receiving an additional quantity of the item after an order for the additional quantity of the item is placed, wherein the reorder time includes a lead time plus one day to account for sales occurring during a reorder day; and
determining a reorder point based on the determined maximal sales estimate for the reorder time, wherein the reorder point indicates a minimum quantity of packages at the particular retail store for maintaining the out-of-stock probability of the item at the particular retail store at or below the acceptable out-of-stock probability.
2. The method according to claim 1 , wherein determining the maximal sales estimate comprises determining the maximal sales estimate using a probability distribution.
3. The method according to claim 2 , further comprising selecting a type of the probability distribution based on a distribution of a quantity of packages per shopper per day, wherein the quantity of packages per shopper per day is based at least on the quantity of packages of the item sold during the period of time and the quantity of shoppers per day that purchased the item.
4. The method according to claim 2 , wherein the probability distribution comprises a Negative Binomial distribution.
5. The method according to claim 2 , wherein the probability distribution comprises a compound Poisson distribution.
6. The method according to claim 5 , wherein a negative binomial distribution, is used as an approximation of the compound Poisson distribution.
7. The method according to claim 1 , wherein an expected daily sales is determined from the sales data and the expected daily sales is used in determining the quantity of shoppers per day.
8. The method according to claim 1 , wherein said determining the maximal sales estimate comprises determining a maximal daily sales estimate such that a probability that the maximal daily sales estimate is exceeded for any given day is at or below the acceptable out of stock probability.
9. The method according to claim 8 , wherein said determining the reorder point based on the determined maximal sales estimate for the reorder time comprises multiplying the maximal daily sales estimate by the lead time plus one day of the reorder time in order to determine the maximal sales estimate for the reorder time.
10. A system for managing inventory comprising:
a computing system configured to execute instruction in order to cause the computing system to perform operations comprising:
receiving a selection of an acceptable out-of-stock probability for an item, the acceptable out-of-risk probability indicating an acceptable risk of the item going out-of-stock at a particular retail store;
accessing sales data for the item, wherein the sales data is for a period of time;
determining a quantity of shoppers per day that purchased the item based at least partly on the sales data;
determining a maximal sales estimate indicating a volume of sales that will cause an out-of-stock probability for the item to exceed the acceptable out-of-stock probability, wherein the maximal sales estimate is determined based on a probability model and at least the quantity of shoppers per day and a reorder time for receiving an additional quantity of the item after an order for the additional quantity is placed, wherein the reorder time includes a lead time plus one day to account for sales occurring during a reorder day; and
determining a reorder point based on the determined maximal sales estimate for the reorder time, wherein the reorder point indicates a minimum quantity of packages at the particular retail store for maintaining the out-of-stock probability for the item at the particular retail store at or below the acceptable out-of-stock probability.
11. The system of claim 10 , wherein the probability model comprises a negative binomial model.
12. The system of claim 10 , wherein the probability model comprises a compound distribution model.
13. The system of claim 10 , wherein a parameter of the probability model is multiplied by the reorder time to account for multi-day sales.
14. The system of claim 10 , wherein the maximal sales estimate is determined for a period of time comprising multiple days.
15. A computer-readable medium having stored thereon instructions that, in response to execution by a computing system, cause the computing system to perform operations comprising:
accessing a selection of an acceptable out-of-stock probability for an item indicating an acceptable risk of the item going out-of-stock at a particular location; and
determining a reorder point based at least partly on a maximal sales estimate for a reorder time, wherein the reorder point indicates a minimum quantity of packages at the particular location for maintaining an out-of-stock probability at the particular location at or below the acceptable out-of-stock probability;
wherein the maximal sales estimate indicates a volume of sales that will cause the out-of-stock probability for the item to exceed the acceptable out-of-stock probability, wherein the maximal sales is determined based on a probability model and at least a quantity of shoppers per day and the reorder time for receiving an additional quantity of the item after an order for the additional quantity is placed, wherein the reorder time includes a lead time plus one day to account for sales occurring during a reorder day.
16. The computer-readable medium of claim 15 , wherein the probability model comprises a negative binomial model.
17. The computer-readable medium of claim 15 , wherein the probability model comprises a compound distribution model.
18. The computer-readable medium of claim 15 , wherein a parameter of the probability model is multiplied by the reorder time.
19. The computer-readable medium of claim 15 , wherein the reorder time accounts for a day the order is placed.
20. The computer-readable medium of claim 15 , wherein the maximal sales estimate is determined for a period of time comprising multiple days.
21. The computer-readable medium of claim 15 , wherein the location comprises a retail store.
22. A method comprising:
accessing a selection of an acceptable out-of-stock probability for an item, the acceptable out-of-stock probability indicating an acceptable risk of the item going out-of-stock at a particular retail store;
accessing sales data for the item, wherein the sales data is for a period of time;
determining, using a computing device, a quantity of packages of the item sold during the period of time based at least partly on the sales data;
determining, using the computing device, a maximal sales estimate indicating a volume of sales that will cause an out-of-stock probability for the item to exceed the acceptable out-of-stock probability, wherein the maximal sales estimate is based on at least the determined quantity of packages of the item sold during the period of time, the acceptable out-of-stock probability, and a reorder time for receiving an additional quantity of the item after an order for the additional quantity of the item is placed, wherein the reorder time includes a lead time plus one day to account for sales occurring during a reorder day; and
determining a reorder point based on the determined maximal sales estimate for the reorder time, wherein the reorder point indicates a minimum quantity of packages at the particular retail store for maintaining the out-of-stock probability of the item at the particular retail store at or below the acceptable out-of-stock probability.
23. The method according to claim 22 , wherein the maximal sales estimate is determined using a probability distribution.
24. The method according to claim 23 , wherein the probability distribution comprises a negative binomial distribution.
25. The method according to claim 23 , wherein the probability distribution comprises a compound Poisson distribution.