System and method for dynamically and automatically updating item prices on e-commerce platform

Patent metadata

Field	Value
Publication number	US-2024338721-A1
Application number	US-202318132190-A
Country	US
Kind code	A1
Filing date	Apr 7, 2023
Priority date	Apr 7, 2023
Publication date	Oct 10, 2024
Grant date	—

Abstract

Systems and methods for dynamically and automatically updating item prices on e-commerce platform are disclosed. In some embodiments, an item is offered for purchase with a current price on a website. When price elasticity data and predicted demand data for the item are both available, a first markdown price is generated for the item using a first model based on: the price elasticity data, the predicted demand data, and the current price. When the price elasticity data and the predicted demand data are not both available, a second markdown price is generated for the item using a second model based on: a decay rate, the current price, and availability of the predicted demand data. A bounded price is generated by applying an upper bound and a lower bound to either the first markdown price or the second markdown price; and transmitted to a computing device for updating the current price of the item on the website.

First claim

What is claimed is:

1 . A system, comprising:
a non-transitory memory having instructions stored thereon; and at least one processor operatively coupled to the non-transitory memory, and configured to read the instructions to:
identify an item being offered for purchase with a current price on a website, wherein the item is desired to reach a target inventory with a price markdown,
determine whether price elasticity data is available for the item,
determine whether predicted demand data is available for the item,
when the price elasticity data and the predicted demand data are both available for the item, generate a first markdown price for the item using a first model based on: the price elasticity data, the predicted demand data, and the current price of the item,
when the price elasticity data and the predicted demand data are not both available for the item, generate a second markdown price for the item using a second model based on: an exponential decay rate, the current price of the item, and availability of the predicted demand data,
generate a bounded price by applying an upper bound and a lower bound to either the first markdown price or the second markdown price, and
transmit the bounded price to a computing device for updating the current price of the item on the website.

2 . The system of claim 1 , wherein:
the computing device is associated with a web server hosting the website; and the item is to be displayed with the bounded price on a webpage of the website, wherein the webpage includes at least one of: a home page of the website, a grocery page including grocery items, an item page including an anchor item, or a promotion page including seasonal or holiday deals.

3 . The system of claim 1 , wherein the at least one processor is further configured to read the instructions to:
obtain inputs from a user device via an integrated interface, wherein the inputs indicate the following information regarding the item: an identification of the item, a start date of a markdown period, an end date of the markdown period, the target inventory, a floor price, and a cadence indicating a minimum quantity of days between two consecutive attempts to reduce price of the item.

4 . The system of claim 3 , wherein the at least one processor is further configured to read the instructions to:
extract the item from a markdown pipeline; determine whether the item has reached the target inventory; determine whether a current time has reached the end date of the markdown period; stop a markdown process for the item when the item has reached the target inventory or the current time has reached the end date; when the item has not reached the target inventory and the current time has not reached the end date, determine whether the current price can be changed based on: the cadence, the current time, and a previous date when the item's price was changed last time; and when the current price cannot be changed, put the item back into the markdown pipeline without price update.

5 . The system of claim 4 , wherein:
it is determined whether the price elasticity data is available for the item when the current price can be changed; and it is determined whether the predicted demand data is available for the item when the current price can be changed.

6 . The system of claim 3 , wherein the at least one processor is further configured to read the instructions to:
collect, for the item, data related to transactions of the item via the website; construct features for the item based on the collected data; and feed the constructed features into a light gradient-boosting machine learning model to generate the predicted demand data for the item, wherein predicted demand data includes a predicted sales rate for the item.

7 . The system of claim 6 , wherein the at least one processor is further configured to read the instructions to:
obtain historical features of a plurality of items offered for purchase on the website, wherein the historical features include at least one of: date and holiday related features, lagged sales features, inventory features, division features, and aggregated features generated from a plurality of most recent markdowns; and train the light gradient-boosting machine learning model based on the historical features.

8 . The system of claim 6 , wherein the first markdown price is generated based on:
computing a desired sales rate for the item based on: a current inventory of the item, the target inventory of the item, and a number of days left before the end date; computing a decay rate for the item based on: the desired sales rate, the predicted sales rate, and the price elasticity data; and computing the first markdown price based on the current price and the decay rate.

9 . The system of claim 6 , wherein the second markdown price is generated based on:
when the predicted demand data is available, determining whether the predicted demand data indicates that the predicted sales rate is enough for the item to reach the target inventory by the end date; computing the second markdown price based on the current price when the predicted sales rate is enough for the item to reach the target inventory by the end date; and when the predicted sales rate is not enough for the item to reach the target inventory by the end date or when the predicted demand data is not available,
computing the exponential decay rate based on: the floor price, the current price, the cadence, and a number of days left before the end date, and
computing the second markdown price based on the current price and the exponential decay rate.

10 . The system of claim 1 , wherein the at least one processor is further configured to read the instructions to:
computing the upper bound based on a minimum of (a) the current price and (b) a universal upper bound; and computing the lower bound based on a maximum of (c) a discounted price by applying a maximum allowed discount on the current price and (d) a universal lower bound.

11 . A computer-implemented method, comprising:
identifying an item being offered for purchase with a current price on a website, wherein the item is desired to reach a target inventory with a price markdown; determining whether price elasticity data is available for the item; determining whether predicted demand data is available for the item; when the price elasticity data and the predicted demand data are both available for the item, generating a first markdown price for the item using a first model based on: the price elasticity data, the predicted demand data, and the current price of the item; when the price elasticity data and the predicted demand data are not both available for the item, generating a second markdown price for the item using a second model based on: an exponential decay rate, the current price of the item, and availability of the predicted demand data; generating a bounded price by applying an upper bound and a lower bound to either the first markdown price or the second markdown price; and transmitting the bounded price to a computing device for updating the current price of the item on the website.

12 . The computer-implemented method of claim 11 , further comprising:
obtaining inputs from a user device via an integrated interface, wherein the inputs indicate the following information regarding the item: an identification of the item, a start date of a markdown period, an end date of the markdown period, the target inventory, a floor price, and a cadence indicating a minimum quantity of days between two consecutive attempts to reduce price of the item.

13 . The computer

Assignees

• Walmart Apollo Llc

Inventors

• Zhang Huidong
• Sarpal Akshit
• Silva Elliot Grenn
• Fu Fangjing
• Huang Fangping
• Song Yang
• Ren Weihang
• Wan Lijie

Classifications

• G06Q30/0283

  Price estimation or determination · CPC title

• G06Q30/0223

  based on inventory · CPC title

• G06Q10/087

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

• G06Q30/0206 Primary

  Price or cost determination based on market factors · CPC title