Automated Batch Picking System with Image Recognition

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing picking systems under the 'goods-to-man' principle face challenges with high picking errors, increased cycle times, and reduced productivity due to the need for manual counting and handling of small items across multiple order containers, often resulting in ergonomic strain and inefficiencies.

Innovation Solution

Implementing an automated system where load carriers and storage aids are exchanged automatically, with a control device managing the counting and delivery processes, ensuring only one source is accessed and a single item is handled at a time, using overhead conveyor technology with pockets to streamline the picking process and reduce errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If manual batch picking is performed with multiple order containers at the picking station, then several customer orders can be processed in parallel, but picking errors increase due to manual counting and handling of small items

Engineering Contradiction:

Improveparallel order processing capacityVSAvoidpicking accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent replaces manual counting and verification with an automated image recognition system using cameras and machine learning algorithms. The system automatically detects, counts, and verifies items placed in order containers, eliminating manual counting errors while maintaining parallel batch picking operations.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system implements real-time feedback through image capture and analysis, providing immediate verification of picked items. The machine learning model continuously learns from feedback to improve accuracy, and the system can alert operators to picking errors or missing items in real-time during batch processing.

Inventive Principle:

Principle #23Feedback

2Productivity

If multiple storage containers and order containers are made available simultaneously at the picking station, then batch picking efficiency is improved, but the work area size increases requiring more walking distance

Engineering Contradiction:

Improvebatch picking throughputVSAvoidpicker walking distance

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The patent transitions from horizontal arrangement of multiple containers to a vertical stacking configuration. Storage containers and order containers are arranged in vertical layers, allowing the picker to access multiple containers within a compact vertical footprint, thereby reducing horizontal work area and walking distance while maintaining batch processing capability.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Productivity

If the picker manually counts and distributes small piece goods to multiple order containers, then batch picking can be performed, but cycle time increases due to repeated counting and container switching

Engineering Contradiction:

Improvebatch picking capabilityVSAvoidpicking cycle time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system enables continuous picking actions by eliminating the need for manual counting and verification between containers. The automated image recognition system continuously monitors and verifies items as they are placed in order containers, allowing the picker to maintain a continuous rhythm without stopping to count or switch containers frequently, thereby reducing cycle time.

Inventive Principle:

Principle #20Continuity of useful action

4Reliability

If manual acknowledgment of source and destination is required during picking, then picking errors can be reduced, but the complexity of the picking process increases

Engineering Contradiction:

Improvesource-destination matching accuracyVSAvoidpicking process complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system implements self-verification through automated image recognition and machine learning algorithms that automatically identify items, count them, and verify they are placed in the correct order containers. This eliminates the need for manual acknowledgment by the picker, reducing process complexity while maintaining or improving accuracy through automated intelligent verification.

Inventive Principle:

Principle #25Self-service

Data Source

PatentEP2766285B2System and procedure for manual batch-picking using the goods-to-man principle

Publication Date: 2022.04.06 SSI SCHAEFER AUTOMATION GMBH (DE)

Figure 1
Figure 2
Figure 3

AI summary

Disclosed is a picking system (10) comprising: an article warehouse area (12), in which a plurality of different articles (84) are stored, properly sorted, in warehouse loading accessories (34); and a picking station (16), wherein the picking station (16) has a first monitoring device (62-1), a second monitoring device, (62-2) and a control device (24) which is connected to the first monitoring device (62-1) and the second monitoring device (62-2) for communication. The control device (24) is adapted to process multiple customer orders (30) according to article, in such a way that warehouse loading accessories (34) can be transported from the article warehouse area (12) to the picking station (16) for the purpose of batch-picking, and wherein the control device (24) is adapted for following a number of article removals from the warehouse loading accessories (34) provided at a removal position (54) in order to effect the exchange of warehouse loading accessories at the removal position (54) and the exchange of a load carrier at each delivery position (58).