QUALITY CONTROL

AND ASSURANCE
(TI 234421)

WEEK #10
MANUFACTURING SYSTEMS LABORATORY
DEPARTMENT OF INDUSTRIAL ENGINEERING
©2024
INTRODUCTION
Quality
Engineering

Off –line QC On-line QC

Design of Information
experiments Feedback

Identify sources Design Diagnostic & Prediction &

of variation optimization adjusment correction

Product Process Process

design design control
EXPERIMENT DESIGN PHASE

To achieve successful experiment requiring four phases :

• Planning the experiment
• Designing the experiment
• Conducting the experiment
• Analyzing the experiment
Design Of Experiment (DOE) is a systematic approach to investigating a
system or process
WHAT IS EXPERIMENTAL DESIGN?
• A test or series of tests in which purposeful changes are made to the input variables of a process so
that we may observe and identify corresponding changes in the output response.

• It may be used for:

– process development
– process troubleshooting to improve process
performance
– obtaining a process that is robust or insensitive
to external sources of variability
THE OBJECTIVES OF THE EXPERIMENT

✓ Determining which variables are most influential on the response, y.

✓ Determining where to set the influential x’s so that y is near the nominal requirement.
✓ Determining where to set the influential x’s so that variability in y is small.
✓ Determining where to set the influential x’s so that the effects of the uncontrollable
variables z are minimized.
EXPERIMENTAL DESIGN APPLICATION

✓ It is a critically important engineering tool for improving a manufacturing process.

✓ It also has extensive application in the development of new processes.
✓ Application of these techniques early in process development can result in:
• Improved yield
• Reduced variability and closer conformance to the nominal
• Reduced development time
• Reduced overall costs
DOE COMPONENTS

• Factors → Factor atau input dari proses dapat diklasifikasikan sebagai variable-variabel yang
dapat dikontrol atau tidak dapat dikontrol.

• Levels → Level atau setting (pengaturan) dari setiap faktor dalam studi yang dilakukan.

• Response → Response adalah output dari proses.

FAKTOR DAN LEVEL
Panjang Batang Tanaman (cm)
Jumlah Air
Ujicoba 1 Ujicoba 2 Ujicoba 3 Ujicoba 4 Ujicoba 5
Disiram (mL)
10 7 7 15 11 9

25 12 17 12 18 18

50 14 18 18 19 19

75 19 25 22 19 23

100 7 10 11 15 11

▪ Faktor : Jumlah Air Variabel Bebas

▪ Level : 10 mL, 25 mL, 50 mL, 75 mL, dan 100 mL 1 Faktor = One-Way ANOVA
▪ Variabel Terikat : Panjang Batang Tanaman
FAKTOR DAN LEVEL
Panjang Batang Tanaman (cm)
Jumlah Air
Ujicoba 1 Ujicoba 2 Ujicoba 3 Ujicoba 4 Ujicoba 5
Disiram (mL)
10 7 7 15 11 9

25 12 17 12 18 18

50 14 18 18 19 19

75 19 25 22 19 23

100 7 10 11 15 11

▪ Faktor A : Jumlah Air ▪ Faktor B : Jumlah Pupuk

▪ Level A : 10 mL, 25 mL, 50 mL, 75 mL, dan 100 mL ▪ Level B : 100 gram, 200 Gram, dan 300 gram
▪ Variabel Terikat : Panjang Batang Tanaman ▪ Variabel Terikat : Panjang Batang Tanaman
CASE STUDY EXAMPLES

1. Characterizing a Process
2. Optimizing a Process
3. A Product Design Example
4. Determining System and Component Tolerances
GUIDELINES FOR DESIGNING EXPERIMENTS
FACTORIAL EXPERIMENTS

• Is used when there are several factors of interest.

• It is the experiment of all possible combinations of the levels of the factors being investigated.
• The effect of a factor is defined as the change in response produced by a change in the level of
the factor (called main effect).
• The main effect of factor A is the difference between the average response at the high level of A
and the average response at the low level of A, or

30 + 40 10 + 20
A = y A+ + y A - = - = 20
2 2
20 + 40 10 + 30
B = y B+ + y B- = - = 10
2 2
NO INTERACTION BETWEEN FACTORS

• At the low level of factor B, the A effect is: A = 30 – 10 = 20

• At the high level of factor B, the A effect is: A = 40 – 20 = 20

40
30
20
10
INTERACTION BETWEEN FACTORS

• In some experiments, the difference in

response between the levels of one factor is not
the same at all levels of the other factors.
• When there is interaction between A and B, the
effect of A depends on the level chosen for
factor B.
• When an interaction is large, the corresponding
main effects have little meaning.

A = y A+ + y A - 30 Low Level
30 + 0 10 + 20 20
= - =0 10
2 2
0 High Level
THE IMPORTANCE OF FACTORIAL EXPERIMENTS
• An alternative to the factorial design that is (unfortunately) used in practice is to change the factors
one at a time rather than to vary them simultaneously.
• For example: first we fix the temperature at 155 oF (the current)

0.5 hr 140 oF

1.0 hr Optimum at 150 oF Optimum at

155 oF 1.5 hr 155 oF 1.7 hr 160 oF 155 oF

1.7 hr 1.7 hr
2.0 hr 170 oF

2.5 hr 180 oF

• The one-factor-at-a-time method has failed here because it fails to detect the interaction between
temperature and time
• Factorial experiments are the only way to detect interactions.
STATISTICAL ANALYSIS OF TWO FACTORIAL EXPERIMENT
THE MODEL
• The observations from a two-factor factorial experiment may be described by the model:

• where μ is the overall mean effect

• τi is the effect of the ith level of factor A
• βj is the effect of the jth level of factor B
• (τβ)ij is the effect of the interaction between A and B
• eijk is an NID(0,σ2) random error component.

• We are interested in testing the hypotheses of no significant factor A effect, no significant

factor B effect, and no significant AB interaction.
SUM OF SQUARES

• The analysis of variance decomposes the total corrected sum of

squares:
SST =SSA +SSB +SSAB +SSE

• The corresponding degree of freedom decomposition is

abn − 1 = (a − 1) + (b − 1) + (a − 1)(b − 1) + ab(n − 1)
 THE ANOVA TABLE FOR A TWO-FACTOR FACTORIAL,
FIXED EFFECTS MODEL

We reject the corresponding hypothesis if the computed F exceeded the tabular value at an appropriate
significance level, or alternatively, if the P-value were smaller than the specified significance level.
THE ANOVA SUM OF SQUARES FORMULA
EXAMPLE
• Aircraft primer paints are applied to aluminum surfaces by two methods—dipping and
spraying.
• A team using the DMAIC approach has identified three different primers that can be used with
both application methods.
• The 18 runs from this experiment were run in random order with the result shown in the table.
• Use the ANOVA to analyze the aircraft primer paint experiment.
SOLUTION
SOLUTION IN SUMMARY

• The P-values for both main effects are very small, indicating that the type of primer used and the
application method significantly affect adhesion force.

• The P-value for the interaction effect F-ratio is relatively large, we would conclude that there is no
interaction between the two factors.

• Conclusion about the hypotheses can also be made using F value from F table → F0.05,2,12 = 3.89
and F0.05,1,12 = 4.75
ANOVA OUTPUT FROM MINITAB
ADHESION FORCE VS. PRIMER TYPE
THE 2K FACTORIAL DESIGN

• The simplest type of 2k design is the 22 → that is, two factors A and B,
each at two levels.
MAIN EFFECTS

• To estimate the main effect of A, we would average the observations on the right side of
the square when A is at the high level and subtract from this the average of the
observations on the left side of the square where A is at the low level, or

• The quantities in the brackets is called contrast.

SIGNS FOR EFFECTS IN THE 22 DESIGN
THE SUM OF SQUARES
EXAMPLE
• Let’s consider two factors: bit size (A) and speed (B). Two levels were chosen for
each factor (bit size at 1/16 inch and1/8 inch and speed at 40 rpm and 80 rpm),
and a 22 design was set up.

• Analyze which factors are important, then perform a statistical analysis and draw a
conclusion.
SOLUTION
Main effects estimation:
ANALYSIS PROCEDURE FOR FACTORIAL DESIGN

Estimate the factor effects

Form preliminary model

Test for significance of factor effects

Analyze residuals

Refine model, if necessary

Interpret results
REGRESSION MODEL
• For the router experiment, the regression model is
y = b0 + b1x1 + b2 x2 + b12 x1x2 + e
• The fitted model is
æ Aö æBö æ AB ö
y = 23.8 + ç ÷ x1 + ç ÷ x2 + ç ÷ x1 x2
è2ø è2ø è 2 ø
æ 16.64 ö æ 7.54 ö æ 8.71 ö
y = 23.8 + ç ÷ x1 + ç ÷ x2 + ç ÷ x1 x2
è 2 ø è 2 ø è 2 ø

• This model can be used to obtain the predicted values of vibration

level at any point in the region of experimentation, including the four
points in the design. For example, for x1 = −1 and low speed x2 = −1;
the predicted vibration level is:

æ 16.64 ö æ 7.54 ö æ 8.71 ö

y = 23.8 + ç ÷ (-1) + ç ÷ (-1) + ç ÷ (-1)(-1) =16.1
è 2 ø è 2 ø è 2 ø
 RESIDUAL ANALYSIS FOR BIT SIZE AND
SPEED PROBLEM

• The residual can be calculated as follow:

RESIDUAL PLOT, WHAT TO CHECK AND WHY?

• To check for independence (response variables

yi are independent), normality, and same
variance.

• Residuals should vary randomly around zero and the spread of the residuals should be
about the same throughout the plot (no systematic patterns).
• If the plot shows a straight line, it is reasonable to assume that the observed sample comes
from a normal distribution.
INTERPRETATION

• Since both factors A (bit size) and B (speed) have large, positive effects, we could reduce
vibration levels by running both factors at low levels.
• Low speed would decrease productivity.
• If we run with speed high and use the small bit, the production rate will be satisfactory.
CONCLUSION

• Design of Experiments (DOE) is one of the statistical approaches in quality

methods applied to enhance service quality in manufacturing and service
industries.
WHY DOE BECOMES IMPORTANT?

✓ Providing Alternative Comparisons.

✓ DOE will identify Significant Inputs.
✓ DOE assists in achieving Optimal Process Outputs.
✓ DOE will reduce Variability.
✓ DOE helps minimize, maximize, or target Outputs.
✓ DOE aids in improving process and product quality.
✓ DOE can balance Tradeoffs.
THANK YOU