Advanced Silicon Processing & Manufacturing Techniques

Mission Statement
About this Postgraduate Programme
University Partners
Industry Partners
Academic and Technical Content
Frequently Asked Questions
How to Apply


Module Number: 5

Title: Optimisation of Processes

Delivered by: University of Edinburgh

Module Credits: 15

Assessment Weighting:

  • Pre-residential work: 3
  • Post-residential work: 7
  • Examination: 10


  • Professor Anthony Walton, University of Edinburgh


  • Professor Anthony Walton
  • Dr Gerard Allan


  • Dr Stewart Smith


  • Dr Christopher Hess, PDF Solutions
  • Dr Andrew Skumanich, Applied Materials
  • Dr Ian Cox, SAS
  • Dr Tim Crandle, Stone Pillar
  • Dr Martin Fallon, National Semiconductor
  • Dr Malcolm Moore,Consultant Engineer

Industrial Advisors:

  • Dr Martin Fallon, National Semiconductor, Greenock
  • Dr Ian Cox, SAS Institute, Bucks


The aim of this module is to present the theoretical and practical knowledge that is required to facilitate the optimisation of IC processing parameters. This Includes methodologies for problem solving, development of robust production processes and approaches for the Design of Experiments (DoE). An insight into methods such as data integration, yield management and data mining to help identify and realise improvement opportunities for existing manufacturing processes will also be given.

Learning Objectives:

Upon successful completion of this modules Delegates will have:

  • a sound understanding of the theoretical techniques for process optimisation
  • experience of the methodologies of problem solving
  • gained an awareness of the capabilities of commercial software for optimisation
  • experience of the effect of six sigma on DFM
  • understand screening and RSM experiments
  • familiarity with IC yield and optimisation strategies
  • an appreciation of techniques to help identify and realise improvement opportunities in manufacturing


  • Pre-residential sessions: assignments 15 %
  • Post-residential sessions: assignments 35 %
  • Examination (supervised) 50 %

Background to the Module:

The production of state-of-the-art integrated circuits is a very complex process with many hundreds of process steps. Hence it is not practical to fully optimise processes using ad hoc procedures; a more structured approach is required. This module is designed to equip engineers with the knowledge required for the optimisation of processes and covers theoretical and practical topics that are generally not found in first degree engineering courses.

Pre-Requisite Knowledge:

Participants are expected to have mathematics at Engineering/Science level and an understanding of IC process and device operation.

Delivery & Assignments:

Pre-residential sessions:

A reading list will sent to delegates before the start of the module. Some material will also be made available on restricted access pages on the World Wide Web and participants will be required to undertake a review / exercises related to the course.

Residential week (35 hours contact time):

Laboratory sessions are a major component of this course. These will be performed using the extensive computer network within the School of Engineering and Electronics. Practical work will be complemented with formal lectures and tutorials including case studies.

Post- residential sessions:

Assignments will be given after the conclusion of the residential period, which will make extensive use of the knowledge gained and test the understanding of the Delegates. A written report of 2000-4000 words and advanced tutorial exercises will be set.

Supervised examination

Candidates will answer 3 questions out of 5 during a 2 hour examination


  • Lectures 16 hours
  • Laboratory & Interactive Sessions 19 hours
Time Topic Content
Day 1 Robust processes and strategies for achieving high quality processes:

Introduction to and comparison of the key strategies for identifying opportunities for process improvement and realizing them; key concepts and methods: Cp, Cpk, six sigma, distributions, z-transformations, bench-marking, design margins, dpu, cycle time reduction, and DFM, Seven Tools of Quality Improvement. Two group exercises are used to illustrate the importance of using DFM and achieving six sigma quality.

Lectures: 3 hours
Interactive exercises: 3 hours

Day 2 Process debugging and improvement - Statistical tools of quality improvement and problem solving:

Graphical tools for exploring relationships; statistical comparisons - t-tests, ANOVA, the interpretation of significance levels and confidence intervals; regression analysis - concept of least squares and assessing the goodness of fit through graphical analysis of residuals, R2and adjusted-R2, and analysis of variance; qualifying and improving measurement capability through gauge R&R analysis.

Lectures: 3 hours
Interactive exercises: 3 hours
Guest Lecture / Case study (1 hour):
Data Integration Requirements for Problem Solving

Day 3 Identifying the critical variables with DOE: Traditional experimental approaches vs statistically deigned experiments; Taguchi and how it differs from traditional experimentation; essential prerequisites to successful experimentation: identification of variables and their experimental range, reliability with which responses are measured, randomisation and blocking; design and analysis of screening experiments: Full Factorial, Fractional Factorial, Plackett Burman and Orthogonal Arrays.

Lectures: 3 1/2 hours
Laboratory Sessions: 3 1/2 hours
Guest Lecture / Case study (1 hour): Realising Data Integration
Day 4 Optimising processes with DOE: Optimising the mean performance using response surface designs: Design and Analysis of Full Factorial, Central Composite, Box-Behnken, D-Optimal Designs, optimizing multiple responses.

Lectures: 3 hours
Laboratory Sessions: 3 hours
Guest Lecture / Case study (1 hour): Yield Management
Day 5 Optimising processes with DOE: Dealing with split plots and data aggregation (nesting) random and fixed effects and their impact upon gauge R&R; Optimising for mean and variance: inner and outer array designs and alternative approaches; design Tolerancing: TCAD and other approaches.

Lectures: 3 hours
Laboratory Sessions: 3 hours:
Guest Lecture / Case study (1 hour): Data Mining

Recommended Texts

Background/General Texts:

  • "Design and Analysis of Experiments", D C Montgomery, John Wiley & Sons, 2000, ISBN: 047-131-6490

  • "World Class Quality: Using Design of Experiments to Make It Happen", K R Bhote, AMACOM, 1999, ISBN: 081-44-04278

  • "Quality Engineering in Producton Systems", G Taguchi, E A Elsayed, Thomas Hsiang, McGraw-Hill, 1989, ISBN: 007-062-8300

  • "Design and Analysis of Circuits", DC Montgomery, John Wiley & Sons, ISBN 0-471-52000-4

General Statistics & Data Analysis

  • "Graphical Methods for Data Analysis", J M Chambers, W S Cleveland, B Kleiner, P A Tukey, CRC Press, 1983, ISBN: 041-205-2717

  • "Applied Regression Analysis", 2nd Edition, N R Draper, H Smith, John Wiley & Sons, 1981

  • "Applied Linear Regression Models", 3rd Edition, J Neter, W Wasserman, M H Kutner, C J Nachtsheim, McGraw-Hill, 1996, ISBN: 025-608-601X

Experimental Design

  • "Statistics for Experimenters: An Introduction to Design, Data Analysis & Model Building", G E P Box, W G Hunter, J S Hunter, John Wiley & Sons, 1978, ISBN: 047-109-3157

  • "Empirical Model-Building and Response Surfaces" G E P Box, N R Draper, John Wiley & Sons, 1990, ISBN: 047-181-0339

  • "Evolutionary Operation: A Statistical Method for Process Improvement", G E P Box, N R Draper, John Wiley & Sons, 1998, ISBN: 047-125-5513

  • "Engineering Quality & Experimental Design", D M Grove, T P Davis, Longman Publishing Group, 1992, ISBN: 058-206-6875

  • "Understanding Industrial Designed Experiments", 3rd Edition, S R Schmidt, R G Launsby, Air Academy Press, 1992, ISBN: 096-221-762X

Variance Components

  • "Analysis of Messy Data, Volume 1: Designed Experiments", G A Milliken, D E Johnson, CRC Press, 1992, ISBN: 041-299-0814

  • "Design and Analysis of Experiments", 2nd Edition, D C Montgomery, John Wiley & Sons, 2000, ISBN: 047-131-6490

Data Mining

  • "The Elements of Statistical Learning: Data Mining, Inference, and Prediction", Trevor Hastie, Robert Tibshirani, Springer-Verlag, 2001, ISBN: 038-795-2845

Applications and Related Journal Articles

  • "Preludes to a Screening Experiment: A Tutorial", R Carlson (1992). Chemometrics and Intelligent Laboratory Systems, 14, pp103-114

  • "Some Things Engineers Should Know About Experimental Design", G J Hahn (1977). Journal of Quality Technology, 9, pp 13-20

  • "A Systematic Approach to Planning for a Designed Industrial Experiment" D E Coleman, D C Montgomery (1993) - (with discussions by B H

  • Gunter G J Hahn, P D Haaland, M A O'Connell, R V Leon, A C Shoemaker, Technometrics, 35, 1, pp. 1-27

  • "Gauge Capability and Designed Experiments I: Basic Methods", D C Montgomery, G C Runger (1993-94), Quality Engineering, 6 (1), pp. 115 - 135

  • "Gauge Capability and Designed Experiments II: Experimental Designs Models & Variance Component Estimation", D C Montgomery, G C Runger (1993-94), Quality Engineering, 6 (2), pp. 289 - 305

  • "Computer-Aided Design of Experiments - Some Practical Experiences", R D Snee, (1985), Journal of Quality Technology, 17, 4, pp 222-236

  • "Experiments in Industry: Design, Analysis, and Interpretation of Results", R D Snee, L B Hare, J R Trout (1985), ASQC Quality Press, Milwaukee


NB: Details of module content, timetable and lecturers may be subject to change.

Return to previous page

Module 5


09.00 - 10.00
Introduction Robust processes and strategies
Graphical Tools
Traditional vs statistical experimental approaches
Response Surface Methodology (RSM)
Split Lots
10.00 - 11.00
DFM review
Basic Statistical Tests
Screening experiments, randomisation, blocking, Full Factorials
Design and Analysis of Full Factorial and Central Composite
Data aggregation (nesting) random and fixed effects
11.00 - 11.15
11.15 - 13.00
Interactive Implementation of DFM techniques (exercise)
Exercise on statistical tests and anova
Exercise on the design and analysis of full factorial screening experiments
Exercise on the design and analysis of RSM designs
Exercise on the design and analysis of nested studies
13.00 - 14.00
14.00 - 15.15
Implementation of Six Sigma, including Benchmarking and benefits
Fractional Factorial Plackett Burman Orthogonal Arrays
Box-Behnken and D-Optimal Designs; Optimizing multiple responses
Optimising for mean and variance; inner and outer array designs and alternative approaches
15.15 - 15.30
15.30 - 16.15
Interactive implementation of Six Sigma (exercise)
Gauge R&R analysis
Exercise on the design and analysis of cost effective screening designs
Exercise on the design and analysis of RSM designs using multiple responses
Exercise on design and analysis of experiments for optimising the mean and variance of several responses
16.15 - 17.15
Interactive implementation of Six Sigma (exercise)
Exercise on regression and gauge R&R
Realising Data Integration
Yield Management
Wrap up
Couse Ends
Enquiries and further information from:

Mrs Sandra Peace
IGDS Programme Co-ordinator,
IGDS Office
School of Electronics & Physical Sciences
University of Surrey

Tel +44 (0)1483 686 138
Fax +44 (0)1483 686 139
Return to top of page
Back to the Homepage