Conveyor Package Validation Using AI Image-Based SKU Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pallet loading systems face inefficiencies due to frequent stoppages caused by package height discrepancies, limited ability to handle multiple workers, and inadequate SKU identification, leading to errors in pallet loading processes.
Innovation Solution
A computer-based system utilizing machine learning models and imaging technology to identify SKUs by analyzing package images, comparing them to desired SKUs, and directing packages to the correct pallets, with features to handle indistinguishable SKU sets and adjust pallet loading sequences dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If package height is used for rough SKU validation on the conveyor, then error detection is facilitated, but frequent stoppages occur due to detected height discrepancies
Solution Approach 1:
The patent replaces the mechanical/optical height measurement system with an AI-based image recognition system. Instead of using breakbeam sensors to detect package height, the system uses imaging cameras coupled with machine learning models to identify SKUs visually, eliminating the frequent stoppages caused by height-based validation while maintaining verification accuracy.
Solution Approach 2:
The patent changes the validation parameter from physical height measurement to visual image analysis. By transitioning from measuring a single dimensional parameter (height) to analyzing multiple visual features through AI image recognition, the system achieves more accurate SKU identification without causing conveyor stoppages.
2Measurement precision
If breakbeam sensors are used to detect package location and validate picks, then pick accuracy is monitored, but the system cannot assign multiple workers to pick packages simultaneously
Solution Approach 1:
The patent implements a universal image recognition system that can simultaneously track and validate multiple packages from multiple workers. The AI model processes images from multiple loading zones concurrently, enabling the system to handle multiple workers without requiring separate detection systems for each worker, thus improving system versatility while maintaining measurement precision.
Solution Approach 2:
The patent merges multiple package detection functions into a single AI-based image recognition system. Instead of using separate breakbeam sensors for each worker's packages, the system combines all package validation into one unified visual recognition process, allowing multiple workers to operate simultaneously while maintaining accurate tracking of each package.
3Reliability
If sequential picks are used for packages of different expected heights, then error detection is facilitated, but the loading process efficiency is reduced
Solution Approach 1:
The patent implements periodic validation through image capture at regular intervals as packages travel through the loading zone. Instead of requiring sequential picks based on height, the system periodically captures images of all packages and uses AI to validate each SKU, maintaining error detection capability while allowing parallel processing that improves loading speed.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances pallet loading efficiency by reducing stoppages, improving SKU identification accuracy, and enabling multiple workers to work simultaneously, ensuring accurate and rapid pallet loading.
Implementation Method 1
taking at least one image of a package on a conveyor
Data Source
AI summary
A pallet loading and validation system provides several features that are particularly beneficial in the context of a conveyor product distribution system, such as a pallet loading system, although some features are not exclusive to a conveyor system. In some aspects, the techniques described herein relate to a method for identifying a SKU of a package using a computer system having at least one processor, the method including: (a) taking at least one image of a package on a conveyor; (b) receiving the at least one image in the computer system; and (c) based upon the at least one image, the computer system determining a SKU associated with the package.