Manufacturing Process Analysis Device Time Shift Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing manufacturing process analysis technologies fail to accurately specify the cause of defects due to differences in measurement times between manufacturing condition values and quality values, leading to inaccurate analysis, especially when the time gap is long or influenced by sensor placement and process conditions.

Innovation Solution

A manufacturing process analysis device that computes the strength of invariant fitness between manufacturing condition values and quality values at different shift times, specifies a specific shift time when the fitness criterion is met, and analyzes the process state based on quality and condition values adjusted by this shift time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If time series data are collected at fixed intervals, then data collection is simple, but measurement time differences cause inaccurate analysis

Engineering Contradiction:
Improvedata collection simplicityVSAvoidanalysis accuracy
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The system performs preliminary alignment of time series data by calculating time differences between manufacturing condition measurements and quality measurements, then shifts one of the data sets to compensate for the time offset before performing correlation analysis. This preliminary action ensures accurate causal relationship detection despite different measurement timings.

Inventive Principle:
Principle #10Preliminary action

2Device complexity

If correlation analysis is performed without time shift correction, then analysis process is simple, but causal relations cannot be clarified

Engineering Contradiction:
Improveanalysis process complexityVSAvoidcausal relation information
Core Design Contradiction:
Device complexityVSLoss of information

Solution Approach 1:

The system dynamically adjusts the time alignment between manufacturing condition data and quality data by calculating optimal time shifts based on process knowledge or data characteristics. This dynamic adjustment allows the analysis to adapt to different measurement scenarios while maintaining causal relationship accuracy.

Inventive Principle:
Principle #15Dynamics

3Duration of action of moving object

If time gap between measurements is long, then process coverage is comprehensive, but correlation analysis becomes inaccurate

Engineering Contradiction:
Improveprocess coverage durationVSAvoidcorrelation analysis accuracy
Core Design Contradiction:
Duration of action of moving objectVSMeasurement precision

Solution Approach 1:

The system introduces an intermediate time shift parameter that acts as a mediator between manufacturing condition measurements and quality measurements. By optimizing this intermediate parameter, the system can accurately correlate events that occur at different times, even with long gaps between measurements.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS10788817B2Manufacturing process analysis device, manufacturing process analysis method, and recording medium whereupon manufacturing process analysis program is stored
Publication Date: 2020.09.29 NEC CORP
  • US10788817B2 patent drawing
  • US10788817B2 patent drawing
  • US10788817B2 patent drawing

AI summary

Provided is a manufacturing process analysis device (30), comprising: a computation unit (31) which computes, in a process in which a manufactured object is manufactured, invariant compliance strengths for each shift time for manufacturing condition values (360) and quality values (361) which are measured in time series; a shift time specification unit (32) which derives, as a specified shift time, a shift time for which the invariant compliance strengths satisfy a baseline; and an analysis unit (33) which analyzes the state of the manufacturing process on the basis of the quality value and the manufacturing condition value for the time which is earlier by the specified shift time than the time at which the quality value is measured.