ML-Guided Item Findability Scoring for Warehouse Picking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Items that are difficult to find in warehouses, such as grocery stores, lead to increased time spent by shoppers and pickers, resulting in lost sales and user dissatisfaction, and can cause online systems to miss out on gross transaction value due to incorrectly determining items as out of stock.
Innovation Solution
A trained machine-learning model is used to predict the findability of items, generating findability scores to enhance item location and facilitate efficient picking by providing action signals for pickers and users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If manual item searching is used in warehouses, then pickers can locate items, but time consumption increases and productivity decreases
Solution Approach 1:
The patent replaces manual mechanical searching with an automated machine-learning-based prediction system that predicts item locations and send alerts to pickers, eliminating the need for manual scanning and searching while significantly reducing time consumption and improving productivity
Solution Approach 2:
The patent introduces a machine-learning prediction model as an intermediary between the inventory system and pickers, which processes item data, predicts findability scores, and generates alerts to optimize the item location process without requiring pickers to manually search
2Measurement precision
If automated systems are used to track items, then item location accuracy improves, but system complexity increases
Solution Approach 1:
The patent uses machine-learning models to process and transform various input parameters (item data, location information, historical patterns) into meaningful findability predictions, achieving high measurement precision through sophisticated parameter transformation rather than complex hardware systems
Solution Approach 2:
The patent creates a virtual copy of the physical warehouse environment through data modeling and machine-learning predictions, allowing the system to simulate and predict item locations without requiring complex physical tracking infrastructure throughout the entire warehouse
3Ease of operation
If pickers search for items manually, then they can find items, but labor costs increase
Solution Approach 1:
The patent enables the system to serve itself by automatically predicting item locations and generating alerts without requiring continuous human intervention, allowing pickers to receive automated guidance rather than manually searching, thereby reducing labor resource consumption while maintaining ease of operation
Solution Approach 2:
The patent replaces manual labor with an automated machine-learning system that predicts item locations and communicates with pickers through alerts and notifications, significantly reducing the quantity of labor resources needed while maintaining or improving item findability
4Productivity
If items are marked as out of stock incorrectly, then order fulfillment is simplified, but gross transaction value is lost
Solution Approach 1:
The patent implements a feedback mechanism where the machine-learning model continuously learns from actual item location data and picker feedback to improve its predictions, reducing false out-of-stock markings and ensuring that items are only marked as unavailable when truly out of stock, thereby preventing loss of gross transaction value while maintaining order fulfillment efficiency
Data Source
AI summary
An online system uses a trained machine-learning model to predict hard-to-find items, which may facilitate picking of these items. The online system receives, from one or more devices of one or more pickers, a device of a source, one or more devices associated with one or more users, and/or a computing system associated with a physical receptacle utilized by at least one user for shopping in a location of the source, data with information about an item. The online system applies the trained machine-learning model to output, based on the received data, a findability score for the item indicative of a findability of the item. Based on the findability score, the online system generates and communicates one or more action signals to a device of a picker, the device of the source, and/or a device associated with a user prompting one or more actions in relation to the item.