Manufacturing Line Causal Modeling With Unsensed Event Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately identify causal relationships between machinery in manufacturing lines, especially when abnormalities involve physical phenomena that cannot be sensed as data, making it difficult for unskilled maintenance personnel to detect and address issues effectively.
Innovation Solution
An information processing device that identifies a first causal relationship based on connection and order data between machinery, acquires a second causal relationship between events related to the machinery through user settings, and synthesizes these relationships to incorporate unsensed events, using nodes and edge information to visualize the causal connections.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a causal relationship model is built using only sensed data between machineries, then the model can be automatically constructed, but events that cannot be sensed as data (physical phenomena) cannot be incorporated into the causal relationship
Solution Approach 1:
The patent introduces an intermediary element (event information) that bridges the gap between sensed data and unsensed physical phenomena. Event information acts as a mediator that can represent both data-driven events and unsensed physical phenomena, allowing the causal relationship model to incorporate both automatic sensing results and human knowledge about unsensed events.
Solution Approach 2:
The patent merges two different types of information: causal relationships derived from sensed data between machineries and causal relationships involving unsensed events. By combining these information sources into a unified causal relationship model, the system achieves both automation through data sensing and completeness through inclusion of unsensed physical phenomena.
2Reliability
If skilled maintenance persons rely on experience and intuition to grasp causal relationships, then they can identify true causes of abnormalities, but unskilled maintenance persons cannot perform effective maintenance work
Solution Approach 1:
The patent creates a formalized copy or representation of the causal relationships that skilled maintenance persons have in their experience and intuition. By encoding these relationships into a structured model with events and causal links, the system captures expert knowledge in a form that can be systematically applied and visualized, making it accessible to unskilled personnel.
Solution Approach 2:
The causal relationship model serves as an intermediary that translates implicit expert knowledge into explicit, visualizable information. This intermediary representation bridges the gap between skilled personnel's intuitive understanding and unskilled personnel's need for guidance, enabling effective maintenance work across all skill levels.
3Productivity
If the number of machineries in the manufacturing line increases, then the production capacity improves, but the complexity of grasping causal relationships between all machineries increases
Solution Approach 1:
The patent segments the complex system of multiple machineries into discrete event elements with defined causal relationships. By breaking down the overall system into individual events and their causal links, the model can handle an arbitrary number of machineries without proportionally increasing complexity, as each machinery is represented through standardized event structures.
Solution Approach 2:
The patent changes the parameter representation from direct machinery-to-machinery relationships to event-based causal relationships. This parameter transformation allows the system to scale to any number of machineries while maintaining manageable complexity, as the event-based model abstracts away the combinatorial explosion of direct relationships.
Data Source
AI summary
An information processing device includes a relationship identifier that identifies a first causal relationship based on data of a connection state and an order relationship of machineries that constitute a manufacturing line, the first causal relationship being between the machineries in a process performed on the manufacturing line, a relationship acquirer that acquires a second causal relationship between events related to the machineries in a process performed on the manufacturing line according to user setting, and a relationship synthesizer that generates a third causal relationship obtained by integrating the first causal relationship and the second causal relationship.