Assembly Line Root Cause Analysis via Product Path Tracking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Identifying malfunctioning nodes in assembly lines is challenging due to the complexity of interactions between nodes, where a first node's bad input can erroneously appear as a malfunction in a subsequent node, and often requires expert human analysis for each new product or part.

Innovation Solution

A system and method for root cause analysis that tracks the paths of failed products through the assembly line using tracking sensors and a database, identifying malfunctioning nodes by analyzing failure rates and edge scores, and automatically generating alerts or adjustments to address these nodes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If expert human analysis is used to identify malfunctioning nodes, then accuracy in identifying the true cause can be improved, but the time and cost required increases significantly and must be reestablished for each new product

Engineering Contradiction:
Improveaccuracy in identifying malfunctioning nodesVSAvoidtime required for analysis
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the mechanical system of expert human analysis with an automated computer-based system that uses machine learning models and algorithms to identify malfunctioning nodes. The system automatically collects data from sensors, processes it through trained models, and generates diagnoses without requiring human expert intervention, thereby eliminating the time and cost associated with manual analysis while maintaining or improving accuracy.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs preliminary action by pre-training machine learning models with historical data before deployment. These pre-trained models can immediately begin analyzing new products and assembly lines without requiring re-establishment of analysis frameworks, enabling rapid identification of malfunctioning nodes across different product lines while maintaining consistent accuracy standards.

Inventive Principle:
Principle #10Preliminary action

2Device complexity

If traditional monitoring methods are used, then implementation is simpler, but the ability to accurately track product paths and identify root causes deteriorates

Engineering Contradiction:
Improvesimplicity of implementationVSAvoidability to track product paths
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent introduces tracking sensors as intermediary devices that are integrated into the assembly line nodes. These sensors automatically capture and transmit product path information to the central system, serving as mediators between the physical assembly process and the digital analysis system. This intermediary layer enables precise tracking without significantly increasing operational complexity, as the sensors are seamlessly integrated into existing node structures.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Reliability

If a first node provides bad input to a second node, then the second node may erroneously appear as malfunctioning, but traditional methods cannot distinguish the true source of the problem

Engineering Contradiction:
Improveaccuracy in identifying true malfunctioning nodeVSAvoidcomplexity of analysis system
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system implements feedback mechanisms where each node's output quality is continuously monitored and fed back to previous nodes in the product path. When a malfunctioning node is identified, the system traces the feedback loop backward through the assembly path to identify upstream nodes that may have provided defective input. This feedback-based approach enables accurate identification of true malfunctioning nodes while maintaining manageable system complexity through structured information flow.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent adds another dimension to the analysis by incorporating temporal and spatial tracking information alongside quality metrics. By analyzing product paths across multiple dimensions (time, location, quality parameters), the system can distinguish between nodes that merely transmit defects versus nodes that actually create them, resolving the ambiguity without requiring excessive system complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP3405844B1Methods and systems for root cause analysis for assembly lines using path tracking
Publication Date: 2020.12.30 ROBERT BOSCH GMBH
  • EP3405844B1 patent drawingFigure 1
  • EP3405844B1 patent drawingFigure 2
  • EP3405844B1 patent drawingFigure 3

AI summary

Methods and systems for performing root cause analysis for an assembly line including a plurality of nodes using path tracking. One method includes receiving tracking data identifying a subset of the plurality of nodes processing a product produced by the assembly line and receiving approval data associated with the assembly line identifying whether each product produced by the assembly line fails an approval metric. The method also includes enumerating a plurality of paths through the assembly line based on the tracking data and determining a failure rate for each of the plurality of paths based on the approval data. In addition, the method includes identifying a malfunctioning path included in the plurality of paths based on the failure rate for each of the plurality of paths, identifying a malfunctioning node based on the malfunctioning path, and performing an automatic action to address the malfunctioning node.