Vision-Based Pallet Verification Using Active Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The delivery process from distribution centers to stores is prone to errors and inefficiencies, particularly in verifying the accuracy of product loading and unloading, leading to additional operating costs and delays due to manual counting and verification processes.
Innovation Solution
An improved delivery system utilizing machine learning and computer vision software, combined with serialized RFID/Barcode technology, to validate pallet loading and unloading, ensuring accurate product placement and reducing the need for manual verification by comparing images of SKUs against pick lists before and after delivery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual counting and verification processes are used to ensure product accuracy, then measurement precision is improved, but loss of time increases due to drivers waiting for clerks
Solution Approach 1:
The system performs preliminary verification by capturing images of products during the loading process at the distribution center, before delivery to the store. This advance verification allows the driver to have proof of correct loading, eliminating the need to wait for store clerk verification and reducing driver waiting time while maintaining accuracy.
Solution Approach 2:
The system creates digital copies (images) of the physical products on the pallet during loading. These image copies are then used for verification purposes at both the distribution center and store, replacing the need for manual physical counting and verification by clerks, thus reducing time loss while maintaining measurement precision.
2Manufacturing precision
If clerks physically count and verify each product on pallets, then manufacturing precision is improved, but productivity decreases due to the time-consuming process
Solution Approach 1:
The system replaces the mechanical manual counting process with an automated optical system using cameras and image processing algorithms. The system captures images of the pallet and automatically identifies, counts, and verifies products through computer vision, maintaining high accuracy while dramatically improving productivity by eliminating the time-consuming manual process.
Solution Approach 2:
The system enables self-verification by automatically comparing the captured images against the expected product list without requiring human clerk intervention. The automated system performs the verification function itself, maintaining precision while freeing up clerks to focus on customer service, thus improving overall productivity.
3Reliability
If drivers wait for clerks to check in product, then reliability of product receipt is improved, but loss of time increases significantly
Solution Approach 1:
The system performs verification in advance by capturing and processing images during the loading process at the distribution center. This preliminary verification creates a digital record that can be immediately used for receipt confirmation, eliminating the need for drivers to wait for store clerk verification and maintaining reliability while reducing time loss.
Solution Approach 2:
The system provides immediate feedback by automatically verifying products against the order list and generating verification results in real-time. This instant feedback mechanism allows drivers to confirm delivery accuracy without waiting for manual clerk verification, maintaining reliability while significantly reducing the time required for product receipt.
Data Source
AI summary
A delivery system generates a pick sheet containing a plurality of SKUs based upon an order. A loaded pallet is imaged to identify the SKUs on the loaded pallet, which are compared to the order prior to the loaded pallet leaving the distribution center. The loaded pallet may be imaged while being wrapped with stretch wrap. At the point of delivery, the loaded pallet may be imaged again and analyzed to compare with the pick sheet.


