Allocating a product inventory to an omnichannel distribution supply chain

Patent metadata

Field	Value
Publication number	US-10423923-B2
Application number	US-201615264195-A
Country	US
Kind code	B2
Filing date	Sep 13, 2016
Priority date	Sep 13, 2016
Publication date	Sep 24, 2019
Grant date	Sep 24, 2019

Abstract

A computer implemented method and system of allocating a target commodity product onto an omnichannel distribution system is provided. Historical data related to the target commodity product is retrieved. Data mining is performed on the retrieved historical data to identify patterns therefrom. An omnichannel nominal demand prediction model is developed based on the identified patterns of the data mining for the omnichannel distribution system. A life-cycle demand is forecast. An allocation for the target commodity product based on the omnichannel nominal demand prediction model is created. A worst case scenario of allocation of the target commodity product for the omnichannel distribution system is identified. The allocation is adjusted to prevent the worst case scenario.

First claim

What is claimed is:

1. A computing device comprising:
a processor;
a network interface coupled to the processor to enable communication over a network;
a storage device for content and programming coupled to the processor;
a revenue management software stored in the storage device, wherein an execution of the software by the processor configures the computing device to perform acts comprising:
retrieving historical data related to a target commodity product;
performing data mining on the retrieved historical data to identify patterns therefrom;
developing an omnichannel nominal demand prediction model based on the identified patterns of the data mining for an omnichannel distribution system;
forecasting a life-cycle demand and creating an allocation for the target commodity product based on the omnichannel nominal demand prediction model;
identifying a worst case scenario of demand for allocation of the target commodity product for the omnichannel distribution system; and
adjusting the allocation to prevent the allocation dependent worst case scenario.

2. The computing device of claim 1 , wherein the historical data comprises transaction log (TLOG) data of a commodity product related to the target commodity product.

3. The computing device of claim 2 , wherein the TLOG data comprises:
key performance indicators (KPI's) comprising at least one of: (i) a gross margin, (ii) a sell through percentage, (iii) a sell through rate, (iv) a percentage sold over total inventory, (v) and a revenue of the commodity product; and
relevant factors comprising a zone, a time, and a channel of sale of each item of the commodity product and at least one of: (i) promotional information at the zone and the time of sale, (ii) weather conditions at the zone and time of sale, and (iii) a public event at the zone and time of sale.

4. The computing device of claim 3 , wherein the relevant factors further comprise at least one of: (i) information whether the sale of each item of the commodity product is part of a basket, and (ii) whether the sale of each item is part of a pack of identical commodity products.

5. The computing device of claim 1 , wherein the historical data is also related to one or more products that are purchased in a same basket as the target commodity product, at a rate that is above a predetermined threshold.

6. The computing device of claim 1 , wherein the omnichannel nominal demand prediction model is based on an average demand of a commodity product related to the target commodity product, for each channel, for different zones, and for different time periods over a life-cycle of the related commodity product.

7. The computing device of claim 1 , wherein identifying the worst case scenario of allocation of the target commodity product comprises:
using an upper limit of uncertainty constraints τ at different zone and channel of sale hierarchy levels;
generating an uncertainty set U(τ) as a group of linear inequalities that constrain an uncertainty variable E;
developing a restricted master program (RMP) that includes extreme points of the uncertainty set U(τ);
developing a mini-max subproblem configured to calculate new extreme points of the uncertainty set U(τ);
iteratively solving the RMP and the mini-max subproblem until a convergence is reached; and
simulating a lifecycle demand based on the value of τ.

8. The computing device of claim 7 , wherein developing the RMP that includes the extreme points of the uncertainty set U(t) comprises identifying a worst case scenario of a combination of ship from store (SFS) fulfilment and a lost sales in stores cost.

9. A non-transitory computer readable storage medium tangibly embodying a computer readable program code having computer readable instructions that, when executed, causes a computer device to carry out a method of allocating a target commodity product onto the omnichannel distribution system, the method comprising:
retrieving historical data related to the target commodity product;
performing data mining on the retrieved historical data to identify patterns therefrom;
developing an omnichannel nominal demand prediction model based on the identified patterns of the data mining for the omnichannel distribution system;
forecasting a life-cycle demand and creating an allocation for the target commodity product based on the omnichannel nominal demand prediction model;
identifying a worst case scenario of allocation of the target commodity product for the omnichannel distribution system; and
adjusting the allocation to prevent the worst case scenario.

10. The non-transitory computer readable storage medium of claim 9 , wherein the method is performed at different times during the life-cycle of the target product.

11. The non-transitory computer readable storage medium of claim 9 , wherein the historical data comprises transaction log (TLOG) data of a commodity product related to the target commodity product.

12. The non-transitory computer readable storage medium of claim 11 , wherein the TLOG data comprises:
key performance indicators (KPI' s) comprising at least one of: (i) a gross margin, (ii) a sell through percentage, (iii) a sell through rate, (iv) a percentage sold over total inventory, (v) and a revenue; and
relevant factors comprising a zone, a time, and a channel of sale of each item of the commodity product and at least one of: (i) promotional information at the zone and the time of sale, (ii) weather conditions at the zone and time of sale, and (iii) a public event at the zone and time of sale.

13. The non-transitory computer readable storage medium of claim 12 , wherein the relevant factors further comprise at least one of: (i) information whether the sale of each item of the commodity product is part of a basket, and (ii) whether the sale of each item is part of a pack of identical commodity products.

14. The non-transitory computer readable storage medium of claim 9 , wherein the historical data is also related to one or more products that are purchased in a same basket as the target commodity product, at a rate that is above a predetermined threshold.

15. The non-transitory computer readable storage medium of claim 9 , wherein identifying the worst case scenario of allocation of the target commodity product comprises:
using an upper limit of uncertainty constraints τ at different zone and channel of sale hierarchy levels;
generating an uncertainty set U(τ) as a group of linear inequalities that constrain an uncertainty variable E;
developing a restricted master program (RMP) that includes extreme points of the uncertainty set U(τ);
developing a subproblem configured to calculate new extreme points of the uncertainty set U(τ);
iteratively solving the RMP and the mini-max subproblem until a convergence is reached; and
simulating a lifecycle demand based on the value of τ.

16. The non-transitory computer readable storage medium of claim 9 , wherein developing the RMP that includes the extreme points of the uncertainty set U(t) comprises identifying a worst case scenario of a combination of ship from store (SFS) fulfilment and a lost sales in stores cost.

17. A computer implemented method comprising:
retrieving historical data related to a target commodity product;
performing data mining on the retrieved historical data to identify patterns therefrom;
developing an omnichannel nominal demand prediction model based on the identified patterns of the data mining for an omnichannel distribution system;
forecasting a life-cycle demand and creating an allocation for the target commodi

Assignees

• IBM

Inventors

• Harsha Pavithra
• Subramanian Shivaram

Classifications

• G06Q10/06393

  Score-carding, benchmarking or key performance indicator [KPI] analysis · CPC title

• G06Q10/087 Primary

  Inventory or stock management, e.g. order filling, procurement or balancing against orders · CPC title

• G06Q30/0202 Primary

  Market predictions or forecasting for commercial activities · CPC title

• G06Q10/08726

  for replenishment processing, procedures, or recommendations using forecasting or optimisation · CPC title