Warehouse Material Flow Control Using Real-Time Virtual Safety Zones

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing warehouse management systems with automated machines and conveyor systems pose safety risks to personnel due to collisions and limited visibility, leading to increased costs and reduced efficiency from extensive safety measures.

Innovation Solution

A method that virtualizes the warehouse using image recognition and a central computer model to track the positions and movements of automated machines and personnel, allowing for real-time control of conveyor systems to prevent collisions and ensure safety without reducing system throughput.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If extensive safety measures such as fencing, light barriers, and sensors are implemented to protect personnel from automated machines, then personnel safety is improved, but system complexity and investment costs increase significantly

Engineering Contradiction:
Improvepersonnel safetyVSAvoidsafety system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent merges the safety monitoring function with the existing vision system used for material flow control. The same cameras and image recognition algorithms that track objects and conveyor positions are also used to detect personnel locations and predict collision risks, eliminating the need for separate safety sensors and light barriers.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The vision system serves multiple functions simultaneously: it controls material flow, tracks objects, monitors conveyor positions, and ensures personnel safety. This multi-functionality reduces the overall number of devices needed while maintaining comprehensive safety coverage.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Reliability

If safety systems with limited line of sight are used to protect personnel, then personnel safety is improved, but system throughput and speed are reduced

Engineering Contradiction:
Improvepersonnel safetyVSAvoidsystem throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system performs preliminary safety assessments by continuously predicting potential collision risks before they materialize. The risk prediction module analyzes personnel positions, machine trajectories, and environmental factors in advance to identify potential hazards, allowing the system to take preventive actions without interrupting normal operations.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The safety control is dynamic rather than static. The system adjusts safety measures in real-time based on actual risk levels: when no risks are detected, the system operates at full speed; when risks are predicted, selective speed reductions or stoppages are applied only to affected areas, minimizing impact on overall throughput.

Inventive Principle:
Principle #15Dynamics

3Reliability

If traditional safety systems individually secure limited areas or defined functions, then specific safety risks are addressed, but the overall safety coverage is insufficient and plant performance is impacted

Engineering Contradiction:
Improvearea safety coverageVSAvoidplant throughput
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent transitions from two-dimensional safety zones (fencing and light barriers defining physical boundaries) to three-dimensional risk prediction zones. The system creates virtual safety envelopes around automated machines and personnel based on their movement trajectories, acceleration patterns, and environmental factors, providing comprehensive coverage without physical constraints.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system implements continuous feedback loops where the risk prediction module constantly receives data from the vision system, updates risk assessments, and adjusts control commands in real-time. This closed-loop control ensures that safety measures are applied precisely when and where needed, minimizing disruptions to plant operations.

Inventive Principle:
Principle #23Feedback

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

This approach simplifies system design, reduces safety device requirements, and maintains high operational efficiency by predicting and preventing dangerous situations, allowing for safer mixed environments with reduced need for physical safety barriers and sensors.

Implementation Method 1

Images of the objects, automated machines and personnel to be conveyed in the conveyor system are captured at predetermined short time intervals using sensors. The objects, automated machines and personnel to be conveyed are identified from the captured images in the central computer using image recognition

Methodology Applied
Scientific EffectImage recognition:

Data Source

PatentEP3821309B1Method and system for controlling the material flow of objects in a real warehouse
Publication Date: 2023.09.13 DEMATIC GMBH
  • EP3821309B1 patent drawingFigure 1
  • EP3821309B1 patent drawingFigure 2

AI summary

Method for controlling a conveyor installation of a real warehouse, which installation comprises automated machines and persons, for which purpose the conveyor installation of the real warehouse, including the persons, is virtualized in a central computer, for which purpose a virtual model of the real conveyor installation having the dimensions of the individual conveyor components and the movement parameters thereof, including the actuator properties, is stored, images of the objects to be conveyed, automated machines and persons in the conveyor installation are captured by means of sensors at predefined short intervals of time, the objects to be conveyed, automated machines and persons are identified from the captured images in the central computer by means of image recognition and the positions thereof in the real conveyor installation at the time of image capture are determined, and the virtual model is continuously updated with the identification and position determination of the objects in the central computer, with the result that a virtualized real-time model is generated therefrom, and the real conveyor installation is centrally controlled with the aid of the virtualized real-time model, wherein material flow control commands are generated for the real actuators for controlling the conveying movement of the respective automated machines in order to avoid endangering the persons.