A Systems Approach to Electronic Assembly Optimization

1
A Systems Approach to Electronic
AssemblyOptimization

• L B Newnes, A R Mileham and A Doniavi
• Chris Egbert
• Dec. 8, 2008

2
Function

• To describe a generic systematic model for
  optimizing Electronics Manufacturing
• To put forth a method for improving an entire
  system rather than discrete subsystems
• Focus on the electronics assembly portion of
  electronics manufacturing

3
Importance

• Electronics manufacturing comprises 1/3 of all
  manufacturing in the world! Dr.
  Red
• Increasing the rate of successfully made
  assemblies will reduce wasted resources and
  increase revenues
• Offer manufacturers a comprehensive and overall
  approach to achieve an optimal system and desired
  targets
• Create a model that may predict the behavior of
  an electronics assembly system

4
References

• 1 Brindley, K. Newnes Electronics Assembly
  Handbook, Techniques,
• Standards and Quality Assurance, 1990 (Heinemann
• Newnes, London).
• 2 Groover, M. P. Fundamentals of Modern
  Manufacturing,
• Material, Processes, and Systems, 1996
  (Prentice-Hall,
• Englewood CliVs, New Jersey).
• 3 Edwards, P. R. Manufacturing Technology in the
  Electronics
• Industry, 1991 (Chapman and Hall, London).
• 4 Conway, P. P., Whalley,D. C., Wilkinson, M. and
  Williams,
• D. J. Automated adaptive control of the reow
  soldering of
• electronic assemblies. In Proceedings of the
  21st IEEE
• International Electronic Manufacturing
  Technology Symposium,
• Austin, Texas, October 1997, pp. 229236.
• 5 Doniavi, A., Mileham, A. R. and Newnes, L. B. A
  systems
• approach to modelling in the manufacturing
  environment.
• In Twelfth National Conference on Manufacturing
• Research, Bath, September 1996.
• 6 Doniavi, A., Mileham, A. R. and Newnes, L. B.
  Computer
• integrated systems engineering in an electronics
  manufacturing

• 8 Wu, B. Manufacturing Systems Design and
  Analysis, 1994
• (Chapman and Hall, London).
• 9 Field, S. W. and Swift, K. G. EVecting a
  Quality ChangeÐ
• An Engineering Approach, 1996 (Arnold, London).
• 10 Milton, J. S. and Arnold, J. C. Introduction
  to Probability
• and Statistics, Principles and Application for
  Engineering
• and the Computer Sciences, 1995 (McGraw-Hill,
  New
• York).
• 11 Doniavi, A. Computer aided systems engineering
  approach
• to electronic manufacturing modelling. PhD
  thesis, University
• of Bath, January 1999.

5
Relation to Technical Area

• It is aimed to increase the yield of electronics
  manufacturing systems as a whole
• Focuses more specifically on the electronics
  assembly aspect of electronics manufacturing (as
  opposed to printed circuit board and
  semiconductor device manufacturing)

6
Parameters Defined

• Stage 1 Setting Objectives
• Stage 2 IDEFO Model
• Stage 4 Generic PCR charts
• Stage 10 Levels Aggregation of Overall System

7
Design Defined

• IDEF, integral computer aided manufacturing
  definition language
• PCh, process characteristics
• PCR, process capability ratio
• MOE, measurement operation evaluation
• FMECA, failure mode effect criticality analysis
• Steps 1, 2, 4, and 10 specifically focused on in
  paper

8
Design Defined (cont.)

• The overall EM process was broken down into the
  IDEFO model of Fig 2
• EA was then broken down to the IDEFO model of Fig
  3
• Each subsystem was then broken down into more
  detailed models

9
Design Principle

• Research performed at a company that manufactures
  switch meters
• IDEFO provided a means to examine the inputs,
  outputs, and interdependencies of each process
• Also allows an approach to obtaining various
  static models
• Generated process maps were useful in validating
  results of stage 1 with industry before
  proceeding to stage 2

10
Experimental Equipment

• IDEFO in software format
• SMC Onsertion
• Lead Component Insertion
• Soldering Flow/ Reflow
• Testing
• Packaging

11
Design Principle Application

• The height of soldering paste on the board is one
  aspect that must be controlled to certain
  specifications
• In order to determine whether the process was
  capable of meeting the specifications, a PCR
  analysis was performed using historical
  industrial data
• Cp can be used to quantify the ability of a
  process to meet the specification

• Where alpha 0.05 and n, the number of
  observations 20,
• The confidence interval is found to be 1.24 lt
  Cp lt 2.33
• Cpk is used to measure the realized process
  capability achieved in actual production and has
  the advantage over the Cp ratio in that it can be
  applied to both skewed and normal distributions
• Cpk confidence interval is
• 0.83 lt Cpk lt 1.68

12
Data and Tables

• Tables indicate that each sub-process has a
  significant impact of the overall yield
• Figure 5 shows the proportion of defects, with an
  average of 16
• Results indicate a failure rate of 16, which
  corresponds to industrial data

13
Correlation of Results with Model

• The lower confidence bound a Cpk value of 1.25
  was determined using
• L was found to be 0.91, therefore the process is
  concluded to be not capable

14
Industrial Use

• Optimization of any aspect of Electronics
  Manufacturing
• Printed Circuit Boards
• Semiconductor Devices
• Electronics Assembly
• As well as all sub-processes
• Identification of processes responsible for
  defects

15
Technical Advancement

• Provides a model for predicting manufacturing
  yield
• Allows analysis of each sub-process, in order to
  identify limiting factors
• Uses statistical approach to give added
  confidence to calculated yield

16
Impacted Industries

• Switch Meters
• PLCs
• Personal Electronics
• Computers
• Data Acquisition Systems
• Motherboards
• Virtually any electronic device industry
  (anything that needs to be plugged in)