Box Item Sorting Sequence Using Constraint-Based Picking Order

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In warehousing and logistics, especially with the rise of online shopping, manual sorting is inadequate for efficiently handling the increased types and quantities of items, and automated sorting is necessary to improve efficiency. However, existing systems struggle to determine the correct picking order for items stacked in boxes, leading to potential damage from incorrect picking sequences.

Innovation Solution

An item picking method and apparatus that acquire picking information including item surfaces, points, and box images. This information is used to determine proximity and constraint relationships between items based on tilt angles and heights, allowing for a logical picking order to be established.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated sorting is implemented to improve sorting efficiency, then productivity increases, but the system cannot determine correct picking order for stacked items leading to item damage

Engineering Contradiction:
Improvesorting efficiencyVSAvoidpicking order accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system performs preliminary analysis of the box content image to determine stacking relationships and constraint relationships before generating the picking order. By pre-processing the spatial information and identifying which items constrain others, the system establishes a safe picking sequence in advance, preventing item damage during automated retrieval.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system introduces an intermediate constraint relationship model that mediates between the physical stacking configuration and the picking sequence. This model represents the constraint relationships between items as a data structure that the planning algorithm can process, enabling the system to reason about picking order without direct physical intervention.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Volume of stationary object

If items are stacked in boxes to increase storage density, then volume utilization improves, but the stacking relationship becomes unknown making correct picking order determination difficult

Engineering Contradiction:
Improvestorage densityVSAvoidstacking relationship information
Core Design Contradiction:
Volume of stationary objectVSLoss of information

Solution Approach 1:

The system replaces physical inspection methods with image-based analysis to determine stacking relationships. By using computer vision to process box content images, the system automatically extracts spatial information about item positions and stacking configurations, eliminating the need for manual inspection while maintaining accurate relationship data.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system transforms the physical stacking configuration into parameterized data representing spatial relationships, constraint relationships, and picking sequences. By converting physical attributes into computational parameters, the system can store and process stacking information in a format suitable for automated decision-making.

Inventive Principle:
Principle #35Parameter changes

3Speed

If picking order is not properly determined, then sorting speed is maintained, but item damage occurs due to incorrect picking sequence

Engineering Contradiction:
Improvepicking speedVSAvoiditem damage
Core Design Contradiction:
SpeedVSObject-affected harmful factors

Solution Approach 1:

The system applies preliminary anti-action by pre-identifying constraint relationships between items and determining a picking order that prevents damage before the picking process begins. The planning algorithm generates a sequence that explicitly avoids actions that would cause harm, such as removing constrained items before their supporting items.

Inventive Principle:
Principle #9Preliminary anti-action

Data Source

PatentEP4556409A1Article sorting method and apparatus, and computer-readable storage medium
Publication Date: 2025.05.21 JINGDONG TECH HLDG CO LTD
  • EP4556409A1 patent drawingFigure 1
  • EP4556409A1 patent drawingFigure 2~3
  • EP4556409A1 patent drawingFigure 4(a)~4(b)

AI summary

An article sorting method and apparatus, and a computer-readable storage medium, which relate to the field of warehouse logistics. The article sorting method comprises: acquiring sorting information of a plurality of articles in a goods box, wherein the sorting information comprises sorting surfaces and sorting points of the articles, and an image of the goods box where the articles are located; determining a proximity relationship between the articles according to the image of the goods box where the articles are located; determining a constraint relationship between the articles according to at least one of an inclination angle and a height of each article, and the proximity relationship between the articles, wherein the inclination angle of each article is determined according to the sorting surface and the sorting point of the article, and the height of each article is determined according to the image of the goods box where the article is located; and determining a sorting sequence of the plurality of articles according to the constraint relationship. By means of the article sorting method and apparatus, the success rate of article sorting can be improved.