Configurable Data Processing Pipelines for Predictive Maintenance
Find Innovative SolutionsGenerate 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
Engineering 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
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.
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.
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
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.
Data Source
Figure 1
Figure 2
Figure 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.