Work Center Machine Ranking for Manufacturing Resource Allocation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In manufacturing setups, efficiently directing operations to the right resources is challenging due to the complexity of allocating resources across multiple work centers, often resulting in sub-optimal allocation, increased costs, and machine breakdowns, especially when dealing with numerous variables and diverse operations.

Innovation Solution

A work center controller utilizes machine learning and decision trees to rank machines based on their suitability for specific operations, leveraging operation and part attributes to provide a ranked list of machines, allowing supervisors to select the most appropriate resource, and continuously improves rankings with operation results.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If combinatorial optimization-based job shop scheduling is used to allocate resources, then resource allocation becomes more systematic, but the complexity of the system increases significantly when there are many variables

Engineering Contradiction:
Improveresource allocation efficiencyVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the complex resource allocation problem into multiple decision trees, each handling specific work centers or operation types. This divides the overall complex optimization problem into smaller, more manageable components that can be processed independently and then integrated.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces decision trees as intermediary structures between the raw scheduling data and the final resource allocation decisions. These decision trees act as mediators that process complex variables through a structured framework, transforming the complex optimization problem into a series of simpler decision rules.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Ease of operation

If heuristic methods are used to direct operations to resources, then the system is simple to operate, but resource utilization becomes inefficient leading to defects and breakdowns

Engineering Contradiction:
Improveoperational simplicityVSAvoidresource allocation reliability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent performs preliminary action by pre-training decision trees with historical data and operation characteristics before actual resource allocation. This advance preparation enables the system to make reliable, data-driven decisions during operation without requiring complex real-time computations, thus maintaining operational simplicity while improving reliability.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback mechanisms where the performance and outcomes of resource allocation decisions are continuously monitored and fed back to refine and retrain the decision trees. This feedback loop progressively improves allocation reliability while maintaining the simplicity of the decision-making process during operation.

Inventive Principle:
Principle #23Feedback

3Productivity

If sub-optimal resource allocation occurs, then the system operates with fewer constraints, but productivity is lost and production costs increase

Engineering Contradiction:
Improvemanufacturing productivityVSAvoidproductivity loss
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The patent changes key parameters by using machine learning models that dynamically adjust resource allocation based on multiple variables including machine status, operation characteristics, and historical performance. This parameter-driven approach optimizes productivity by selecting the best resource for each operation based on current conditions rather than fixed rules.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent introduces dynamics into the resource allocation system by using adaptive decision trees that can be retrained and adjusted based on changing conditions and accumulated data. This dynamic approach allows the system to continuously improve its allocation decisions, maximizing productivity and minimizing losses as it adapts to actual plant conditions.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3663997B1Machine learning based resource allocation in a manufacturing plant
Publication Date: 2023.08.09 SAP SE
  • EP3663997B1 patent drawingFigure 1
  • EP3663997B1 patent drawingFigure 2
  • EP3663997B1 patent drawingFigure 3

AI summary

A work center in a manufacturing setup includes a machine learning model that uses a decision tree to facilitate the work of a supervisor on the production line to choose a machine to perform a particular operation on a particular part. The decision tree outputs a ranking of machines indicating the suitability of the ranked machines for performing the particular operation on the particular part.