Read Slide 1 text version

QBASE

Engineering

DESIGN OF EXPERIMENT TAGUCHI METHODS

Overview Taguchi Method refers to the collection of statistical techniques and philosophy advocated by Professor G. Taguchi. This philosophy forms the basis for the modern concept of "off-line quality control". This can be viewed as an evolution of the modern quality improvement technique. In the 30's manufacturers relied mostly on product inspection to improve product quality. This meant that adjustment on the process could only be made on samples of finished products; too little, too late!. Realizing that this method was costly and most of the time ineffective, Statistical Process Control ( SPC) was later introduced in the Japanese manufacturing, twenty years later. They have demonstrated that better quality can be achieved at reduced cost by controlling the manufacturing processes. However, they later found out that even with excellent process control, manufacturing variation will result in some product variations In 1980's, Dr.Taguchi introduced the concept of achieving product quality by better product and process design. He stressed that traditional designers paid little attention to the variation of functional performance from manufacturing variation, and therefore Taguchi advocated the philosophy that thoroughly evaluates a new product or process not just for functional performance but also for its manufacturability i.e. sensitivity to functional variation. Thus the term " robust design " was coined. The technique for implementing this philosophy has been around for some time - statistical method for experimental design, and today it has been widely used in various manufacturing industries in Japan and the United States. This course is designed to deliver Taguchi Statistical Experimental techniques and applications in a simple and practical manner, packed with examples and exercises , and most important of all, it is designed to be easily understood by non-statistician audiences. In-class experimental design simulation exercise utilizing the techniques will be performed by the participants to reinforce their understanding of DETM to enable them to conduct the experiments in their own working environment. Workbooks will be issued and used during the simulation exercise

© QBASE Engineering Sdn Bhd

QBASE

Engineering

Course Objectives

At the conclusion of this training the participants will have the necessary knowledge to be able to: ·Explain the basic of Taguchi Quality philosophy, Taguchi's Loss Function, and its application to determine realistic manufacturing Tolerances and to assess the Loss impact of process centering and variation reduction efforts; ·Explain the weakness of OVAT experiments as compared to statistically designed experiments; ·Understand Taguchi's approaches and strategies to experimental designs. ·Explain the concept of Taguchi's Orthogonal arrays, interactions and linear graphs, signal to noise ratio, control and signal factors, minimum hurdles, center points and etc. ·Differentiate the differences between two-level Orthogonal Arrays and linear graphs and some 3 level Orthogonal Array; ·Explain the Signal to Noise Ratio and off-line Quality Controls; At the conclusion of this training the participants will have the necessary skills to be able to: ·Plan and design experiments using Taguchi's Orthogonal arrays, Linear graph, signal to noise Ratio to identify signal and control factors in a process; ·Plan, design, test, execute, analyze and conclude a statistically valid experiments for the purpose of parameter screening, process improvements, process optimization, tolerance and parameter design and/or root Cause analyses. ·Apply ANOVA and other minimum hurdle tests for both continuous and attribute experimental data to determine the factor `s significance ·Calculate the impact of process optimization in term of Loss Functions.

© QBASE Engineering Sdn Bhd

QBASE

Engineering

Target Audience

·Manufacturing/Process Engineers and Managers ·Design Engineers and Managers ·Product Engineers and Managers ·Quality Engineers and Managers ·TQM Managers ·Technical Staff

Methodology

·Classroom lecture on concepts, methodologies and techniques ·Workshop and simulation excercises ·Case studies , case discussion and best practices ·Learning from actual examples

Prerequisite

·Statistical Process Control Techniques ·Scientific Calculators

Course Duration

·Two(2) days 9:00 ­ 5:00 pm

Delivery

· English and / or Bahasa Malaysia

Fees per Session

· 2 days at RM 1200 per pax. · Contact QBASE for special Discounts and Terms and Conditions ·* Contact QBASE for in-house training fee

© QBASE Engineering Sdn Bhd

QBASE

Engineering

Course Outlines Introduction · The Taguchi Loss Function · Quadratic Loss · K-the Loss Parameter · Manufacturing Tolerances · Lower level Tolerances · M-S-D · Evaluating off-Center Process · One Variable At a Time (OVAT) Experiments · Introduction to Experimental Design · Process Design Simulations · Basic Taguchi's experimental Design · Basic Orthogonal Array (OA) · Effect Table · Interaction Plot · Minimum Hurdle ·Higher Level Orthogonal Arrays · Two Level Designs (L8,L16,L12, L16..etc) ·Linear Graphs · Three Level Designs (L9, L27) · Special Orthogonal Arrays ·Significant Tests · In-class Experiment Simulations · ANOVA · Concept of ANOVA Calculations · ANOVA for Examples · ANOVA for Two and Three Level OA's ·DOE for Attribute Data ·Likert Scale ·Data Categorization ·Omega Transformation ·Accumulation Analysis

· Taguchi Approach to Process Optimization · Control and Signal Factors · Signal to Noise Ratio (SNR)

· Two stage Optimization

· Concept and Application of Parameter/

· Tolerance Design Discussion and Wrap-Up

· Group Project In class DOE Simulation Exercise Using DOE Workbook/software

·Taguchi Experimental Strategies

© QBASE Engineering Sdn Bhd

QBASE

Engineering

Course Agenda Day 1 9:00 ­ 10:00 Module 1: INTRODUCTION Taguchi Loss Functions One Variable at a Time Simulation Exercise (OVAT) Day 2 9:00 10:00 Module 5: TAGUCHI APPROACH TO PROCESS OPTIMIZATION - Control and Signal factors - Signal to Noise Ratio (SNR) - SNR Calculation - SNR Effect Tables - Two Stage Process Optimization Strategy

10:0010:15 10:15 1:00

Morning Break Module 2: INTRODUCTION TO EXPERIMENTAL DESIGN Basic Orthogonal Arrays (OA) L4 Orthogonal Arrays Linear Graphs Effect Table Calculations Interaction Plots Minimum Hurdles Calculation Module 3: OTHER TWO LEVEL ORTHOGONAL ARRAYS (OA)

10:00 - Morning Break 10:15 10:15 Continuation of Module 5: SNR ­ 1:00 Calculations and Examples

1:00 2:00 2:00 3:00

Lunch Break Continuation of Module 3: Other Two Level Orthogonal Arrays and Linear Graphs L8, L16, L12, Afternoon Break Module 4: ANOVA

Concept and Interpretation of ANOVA ANOVA Calculations ANOVA Examples ANOVA for Two and Three Level Designs

© QBASE Engineering Sdn Bhd

1:00 - Lunch Break 2:00 2:00 - Module 6: DOE for Attribute Data 3:00 Likert Scale OMEGA Transformation Effect Calculations & ANOVA 3:003:15 3:155:30 Afternoon Break DOE WORKSHOPS and PROJECT DISCUSSIONS

3:00 3:15 3:25 5:30

Information

Slide 1

5 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

444145


Notice: fwrite(): send of 198 bytes failed with errno=104 Connection reset by peer in /home/readbag.com/web/sphinxapi.php on line 531