Configurable Data Processing Pipelines for Predictive Maintenance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In production facilities and plants, it is challenging to set up appropriate data processing due to fixed processes and orders in software or applications used for predictive maintenance and abnormality detection, leading to suboptimal data analysis settings.

Innovation Solution

An information processing apparatus and method that allow for the definition of processing operations for multiple processes, acceptance of processing patterns including selected processes and their execution order, and generation of processing setting information to execute the defined pattern.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If fixed processes and execution orders are used in software for predictive maintenance and abnormality detection, then the system structure is simple and easy to implement, but the data processing settings become suboptimal and cannot be adapted to specific facility requirements

Engineering Contradiction:
Improvedata processing setting adaptabilityVSAvoidsystem configuration complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the data processing system into multiple independent processes (e.g., data collection, preprocessing, analysis, output) that can be individually selected and configured. Each process is defined separately with its own parameters, allowing users to build customized processing pipelines by combining only the necessary segments for their specific facility needs, thereby achieving adaptability without overwhelming complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic configurability where the processing pattern (selection of processes and their execution order) can be changed according to different facility requirements. The system transitions from fixed to dynamic configuration, allowing users to adjust the data processing pipeline to match specific predictive maintenance or abnormality detection needs while maintaining a manageable structure through standardized process definitions.

Inventive Principle:
Principle #15Dynamics

2Manufacturing precision

If customized data processing patterns are allowed with selectable processes and execution orders, then optimal data analysis can be achieved for specific facilities, but the system complexity increases

Engineering Contradiction:
Improvedata analysis optimizationVSAvoidprocessing pattern configuration
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-defining multiple standardized processes with clear parameters and execution requirements before the actual data processing task. These pre-configured processes (such as data collection, filtering, transformation, analysis) serve as building blocks that users can select and combine, reducing the complexity of customization while enabling optimal data analysis configurations for specific facility requirements.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4517451A1Information processing apparatus, information processing method, and information processing program
Publication Date: 2025.03.05 YOKOGAWA ELECTRIC CORP
  • EP4517451A1 patent drawingFigure 1
  • EP4517451A1 patent drawingFigure 2
  • EP4517451A1 patent drawingFigure 3~4

AI summary

A data processing device (10) stores definition information for defining a processing operation of each of a plurality of execution processes, accepts, as a processing pattern, a selection of each of the execution processes that are used for data processing and a selection of an execution order the selected execution process, and generates, based on the accepted processing pattern and the definition information, processing setting information for causing the processing pattern.