Self-Learning Recipe Generation for Food Processing Adjustment

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing food processing machines require time-consuming and labor-intensive adjustments to adapt to changing machine types and packaging materials, with expert knowledge often conflicting with user interests, and existing systems struggle to optimize operations quickly and precisely.

Innovation Solution

A device with a self-learning process parameter generator and recipe generator that adapts rules using machine learning algorithms, incorporating data from ongoing operations to refine and expand the set of rules, allowing for continuous improvement and alignment with actual operational conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If expert rules are used to determine recipe variables, then the packaging machine can operate with predefined parameters, but the system cannot adapt quickly to changing machine types and packaging materials

Engineering Contradiction:

Improveadaptability to changing machine types and packaging materialsVSAvoidtime-consuming adjustment process

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system performs self-learning by automatically collecting operational data, analyzing it through machine learning algorithms, and updating its own rule set without requiring external expert intervention. This enables the system to adapt to new machine types and packaging materials autonomously, resolving the contradiction between adaptability and time consumption.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system implements continuous feedback loops where operational data is collected from sensors and processing units, analyzed to evaluate the performance of existing rules, and used to refine and update the rule set. This feedback mechanism enables rapid adaptation to changing conditions while maintaining optimized performance.

Inventive Principle:

Principle #23Feedback

2Manufacturing precision

If expert knowledge is used to create rules, then the system can function with limited initial data, but the rules may not align optimally with user interests and actual operational conditions

Engineering Contradiction:

Improveprecision of process parametersVSAvoidcomplexity of rule adaptation system

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system replaces manual expert rule creation and adjustment with automated machine learning algorithms that analyze operational data to generate and refine rules. This substitution of mechanical/expert processes with computational algorithms improves parameter precision while managing system complexity through automation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system dynamically adjusts rule parameters based on analyzed operational data, transforming static expert-defined rules into adaptive parameters that optimize performance for specific machine types and packaging materials. This parameter adaptation enables precise manufacturing while the system learns from actual operational conditions.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If the recipe generator uses fixed rules, then the system structure remains simple, but the system cannot continuously improve or learn from operational data

Engineering Contradiction:

Improveproductivity of packaging machineVSAvoidcomplexity of self-learning system

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system transitions from static fixed rules to dynamic adaptive rules that continuously evolve based on operational data. The rule set becomes a living system that adjusts its parameters and structure over time, enabling continuous improvement of productivity while the system learns from real-world operations.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system implements continuous learning and adaptation through ongoing data collection, analysis, and rule refinement. This continuous useful action ensures the system constantly improves productivity by incorporating new operational insights, transforming the learning process into an ongoing value-generating activity rather than a periodic update.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentEP4089494A1Self-learning food processing device and method

Publication Date: 2022.11.16 MULTIVAC SEPP HAGGENMULLER GMBH & CO KG

Figure 1
Figure 2
Figure 3

AI summary

The invention relates to a device (1) comprising at least one food processing machine (A - F) and a recipe generator (R) configured to adapt a set of rules (RW) based on data acquired at the food processing machine (A - F). Furthermore, the invention relates to a method for generating a recipe data set for a setting process of a food processing machine (A - F), wherein a recipe generator (R) used to produce the recipe data set can be trained on the basis of machine settings and/or measured values acquired during the operation of the food processing machine (A - F).