Evaluation Model Generation Based on Raw Material Similarity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems face increased processing load and time loss due to the frequent generation of new evaluation models for varying crude oil properties, which are machine-learned to evaluate the state of equipment, particularly in oil refineries, as properties of crude oil vary by region and time of production.

Innovation Solution

An evaluation model generating apparatus determines whether to generate a new evaluation model based on the similarity of crude oil properties using machine learning, clustering raw materials, and generating models only when necessary, thereby reducing processing load and time loss.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If evaluation models are frequently generated for varying crude oil properties, then model accuracy for different raw materials is improved, but processing load and time loss increase

Engineering Contradiction:
Improvemodel accuracyVSAvoidprocessing load
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent changes the parameter of model generation frequency from frequent to conditional based on raw material property variations. The determining unit evaluates whether property data variations exceed a threshold before triggering model generation, thus reducing unnecessary processing while maintaining accuracy when needed.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system performs self-evaluation by automatically determining whether new evaluation models are needed based on property data comparisons. The determining unit autonomously assesses property variations and triggers model generation only when necessary, eliminating the need for continuous external intervention.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If evaluation models are frequently generated for varying crude oil properties, then model accuracy for different raw materials is improved, but time loss increases

Engineering Contradiction:
Improvemodel accuracyVSAvoidtime loss
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent changes the parameter of model generation timing from continuous/frequent to conditional/event-driven. By monitoring property data variations and triggering model generation only when changes exceed thresholds, the system reduces time loss while maintaining model accuracy for varying crude oil properties.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The system implements periodic evaluation of property data against existing models, generating new models only when periodic checks reveal significant variations. This periodic action replaces continuous model generation, reducing time loss while ensuring accuracy when raw material properties change.

Inventive Principle:
Principle #19Periodic action

3Productivity

If evaluation models are generated based on raw material properties, then processing load is reduced, but determination complexity increases

Engineering Contradiction:
Improveprocessing loadVSAvoiddetermination complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the model generation process into two distinct parts: a determining unit that evaluates whether model generation is needed, and a model generation unit that creates models only when triggered. This segmentation reduces overall processing load by separating the lightweight determination step from the resource-intensive model generation step.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The determining unit acts as an intermediary between property data input and model generation. It mediates by evaluating property variations and deciding whether to trigger model generation, thus reducing unnecessary processing load while managing determination complexity through a dedicated intermediate component.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4303672B1Evaluation model generating apparatus, evaluation model generating method, and evaluation model generating program
Publication Date: 2026.04.22 YOKOGAWA ELECTRIC CORP
  • EP4303672B1 patent drawingFigure 1
  • EP4303672B1 patent drawingFigure 2
  • EP4303672B1 patent drawingFigure 3

AI summary

SOLUTION: An evaluation model generating apparatus, including: a property data obtaining unit which obtains property data representing a property of a raw material to be used in equipment that manufactures a product from the raw material; a determining unit which determines, based on the property data, whether to generate an evaluation model that outputs an indicator obtained by evaluating a state of the equipment; and an evaluation model generating unit which generates, according to a result of the determining, the evaluation model by machine learning, is provided. In the evaluation model generating apparatus, the determining unit may determine, if it is judged that a target raw material is not similar in property to a raw material that has been used in the past based on the property data, to generate the evaluation model.