Automated Bag Packing Simulation for Deformable Ordered Items

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional packaging systems are inefficient, leading to overuse of materials, damage to items, and increased shipping costs due to improper packaging, particularly in handling items that deform in one or two dimensions when packed in bags.

Innovation Solution

An automated system that iteratively simulates the packaging process to determine optimal packaging for a combination of goods by selecting a data structure representing a bag, simulating the placement of the largest item, and calculating remaining spaces to efficiently pack all items, thereby reducing material usage and shipping costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional volume-based sorting and box packaging is used, then packaging speed is improved, but packaging quality deteriorates leading to damaged goods and poor customer satisfaction

Engineering Contradiction:
Improvepackaging speedVSAvoidpackaging quality
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent replaces manual/mechanical packaging decisions with an AI-based simulation system that computationally determines optimal packaging arrangements. The system uses iterative simulation to model how items fit in different packaging configurations, substituting physical trial-and-error with digital optimization to achieve both speed and quality.

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

Solution Approach 2:

The system performs preliminary simulation and optimization of packaging arrangements before actual packaging occurs. By pre-calculating the optimal configuration using AI algorithms that consider item dimensions, fragility, and bag deformability, the system prepares packaging instructions in advance, ensuring both speed and quality without real-time adjustments.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If manual labor is used for packaging, then packaging quality is improved, but time consumption increases

Engineering Contradiction:
Improvepackaging qualityVSAvoidpackaging time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system enables self-service packaging by generating automated instructions that guide the packaging process. The AI-determined optimal arrangements are presented as clear instructions to workers or automated packaging machines, allowing the system to determine the packaging plan autonomously while reducing reliance on manual expertise and minimizing packaging time.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces manual packaging labor with an AI-based decision-making system that automatically determines optimal packaging configurations. This substitution eliminates the time-consuming nature of manual assessment and arrangement while maintaining or improving packaging quality through computational optimization.

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

3Reliability

If overuse of boxes and packaging materials is used, then packaging reliability is improved, but cost increases

Engineering Contradiction:
Improvepackaging reliabilityVSAvoidpackaging material usage
Core Design Contradiction:
ReliabilityVSLoss of substance

Solution Approach 1:

The system optimizes packaging by dynamically adjusting parameters such as bag size selection, item arrangement orientation, and packaging material allocation based on the specific dimensions and characteristics of the items being packaged. This parameter optimization ensures adequate protection while minimizing material waste by avoiding the conventional approach of using standardized oversized boxes for all items.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent applies local quality optimization by determining packaging requirements specific to each item's characteristics rather than using uniform packaging for all items. The AI system analyzes individual item properties (dimensions, fragility, shape) and determines the precise packaging needs for each, allowing for material efficiency while maintaining protection where specifically required.

Inventive Principle:
Principle #3Local quality

4Loss of substance

If bag packaging is used for deformable items, then material efficiency is improved, but measurement and simulation complexity increases

Engineering Contradiction:
Improvepackaging material efficiencyVSAvoidbag deformation measurement
Core Design Contradiction:
Loss of substanceVSDifficulty of detecting and measuring

Solution Approach 1:

The system creates a digital copy or virtual model of the bag and items to simulate the packaging process. By representing the deformable bag and items in a computational environment with defined parameters, the system can iterate through multiple packaging scenarios and determine optimal arrangements without physically manipulating actual deformable materials, thus reducing measurement complexity.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent manages bag deformation complexity by parameterizing the simulation model with key characteristics such as bag flexibility, item dimensions, and deformation constraints. By transforming the continuous deformation problem into a parameterized optimization problem, the system can efficiently evaluate multiple configurations while accounting for bag deformability without requiring complex real-time measurement systems.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS11327447B2Systems and methods for computer-determined efficient bagging of ordered items
Publication Date: 2022.05.10 COUPANG CORP
  • US11327447B2 patent drawing
  • US11327447B2 patent drawing
  • US11327447B2 patent drawing

AI summary

The present disclosure provides systems and methods for automated bag packaging, comprising at least one processor; and memory storing comprising: receiving an order comprising at least one item; searching at least one data store to determine one or more properties associated with each item; for each group: performing an optimization process for packaging the at least one item into one or more bags, by: selecting a data structure representing a first bag, the data structure comprising a size of the first bag; iteratively simulating packaging of a largest item of the group into the first bag until all items are packaged in the selected bag; generating at least one set of instructions for packaging the items into the selected bag; and sending the generated instructions to a computer system for display, the instructions including at least one item identifier and one bag identifier.