AI Object Routing and Handling for Warehouse Throughput

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional object processing systems, such as automated storage and retrieval systems, face inefficiencies in time and resource usage when moving and processing a large number of objects, particularly in dynamic environments like order fulfillment centers where a wide variety of goods need to be identified and routed.

Innovation Solution

An automated material handling system that employs AI and machine learning to classify objects based on their features and assign them to appropriate processing stations, optimizing routing and handling parameters through a feedback learning process, utilizing programmable motion devices and sensors to improve throughput and accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional automated storage and retrieval systems use totes brought to people for item picking, then the system allows separation of automated transport and manual picking functions, but the system requires significant time and resources to move totes toward and away from each person, reducing overall processing efficiency

Engineering Contradiction:

Improveseparation of automated transport and manual picking functionsVSAvoidtime required to move totes toward and away from each person

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

Instead of bringing totes to stationary pickers, the system inverts the approach by bringing pickers to the totes using mobile robotic devices. This allows the picker to move directly to the tote location, eliminating the time-consuming process of transporting entire totes back and forth while maintaining the separation of automated and manual functions

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The system segments the picking function from the transport function by using mobile robotic devices that can independently navigate to tote locations. This segmentation allows the picker to be mobile rather than stationary, reducing the time spent moving totes while maintaining functional separation

Inventive Principle:

Principle #1Segmentation

2Productivity

If conventional systems require people to process a large number of totes in a fixed location, then the transport system can focus on moving totes efficiently, but the person's ability to process totes quickly is limited by their stationary position and the frequency of tote arrivals

Engineering Contradiction:

Improverate at which person can pick goodsVSAvoidtime spent waiting for totes to be brought to the person

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system inverts the traditional workflow by making the picker mobile rather than stationary. The mobile robotic device transports the picker directly to tote locations, eliminating waiting time and allowing the picker to process totes on demand without being constrained by fixed location and scheduled arrivals

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The system introduces dynamic mobility to the picker through mobile robotic devices. Instead of a static picking station, the picker can dynamically move to different tote locations, optimizing the processing rate by eliminating idle waiting time and adapting to varying tote locations and priorities

Inventive Principle:

Principle #15Dynamics

3Productivity

If automated systems handle object routing and processing, then throughput and accuracy can be improved, but the system requires sophisticated classification and routing capabilities to handle a wide variety of objects with different characteristics

Engineering Contradiction:

Improveobject processing throughputVSAvoidclassification and routing system complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system performs preliminary classification of objects at the supply location using the object classification system before routing. This preliminary action allows objects to be sorted into appropriate classes based on their characteristics, enabling more efficient routing and processing while reducing the complexity of real-time decision-making during processing

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The object classification system acts as an intermediary between the supply location and processing stations. It evaluates object information, determines appropriate processing parameters, and routes objects to suitable stations, thereby managing system complexity by centralizing the classification and routing intelligence in a dedicated intermediary system

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11407589B2Systems and methods for learning to extrapolate optimal object routing and handling parameters

Publication Date: 2022.08.09 BERKSHIRE GREY OPERATING CO INC

AI summary

A system for object processing is disclosed. The system includes a framework of processes that enable reliable deployment of artificial intelligence-based policies in a warehouse setting to improve the speed, reliability, and accuracy of the system. The system harnesses a vast number of picks to provide data points to machine learning techniques. These machine learning techniques use the data to refine or reinforce in-use policies to optimize the speed and successful transfer of objects within the system. For example, objects in the system are identified at a supply location, a predetermined set of information regarding object is retrieved and combined with a set of object information and processing parameters determined by the system. The combined information is then used to determine routing of the object according to an initial policy. This policy is then observed, altered, tested, and re-implemented in an altered form.