BY:

Operational Excellence team

Happy Learning….
Analyze

2
Analyze

Purpose
To pinpoint and verify causes affecting the key input and output
variables tied to project goals. (“Finding the critical Xs”)

Deliverables
• Data charts and other analyses that show the link between the
targeted input and process (Xs) variables and critical outputs (Y)

• Identification of value-add and non-value-add work

• Calculation of process cycle efficiency

3
Analyze
Key Steps
1.Conduct value analysis: Identify value-add and non-value-add
steps

2.Calculate Process Cycle Efficiency (PCE): Compare to world

class benchmarks to help determine how much improvement is
required

3.Analyze Process flow: Identify bottleneck points & constraints in a

process and assess their impact on the process throughput and its
ability to meet customer demands and CTQs

4.Analyze data collected in Measure

4
Analyze
Key Steps
5. Generate theories to explain potential causes: Use
brainstorming, FMEA, Cause & Effect diagrams, and other tools
identify potential causes of the observed effects

6. Narrow the search: Use brainstorming, selection, and

prioritization techniques (Pareto charts, hypothesis testing, etc.) to
narrow the search for root causes and significant cause-and-effect
relationships

7. Collect additional data to verify root causes: Use scatter plots

or other sophisticated statistical tools e.g. hypothesis testing, ANOVA
to verify significant relationships

5
Analyze
Hypothesis Testing
● A statistical hypothesis is an assumption about a population
parameter

● Parameters represent “what we wish to know” about a population.

Statistics are used to estimate population parameters.

● The best way to determine whether a statistical hypothesis is true

would be to examine the entire population.

● Since that is often impractical, researchers typically examine a random

sample from the population.

● If sample data are not consistent with the statistical hypothesis, the
hypothesis is rejected.

6
Analyze
Hypothesis Testing
Hypothesis Testing enables us to:

● Tell us whether or not two sets of data are different

● Tell us whether or not a statistical parameter (average, standard

deviation, etc.) is different from a value of interest

● Allow us to assess the “strength” of our conclusion (our probability of

being correct)

● Control our risk of making wrong decisions or conclusions

7
Analyze
Hypothesis Testing

Two types of statistical hypotheses:

● Null hypothesis: The null hypothesis, denoted by H0, is usually the

hypothesis that sample observations result purely from chance

● Alternative hypothesis: The alternative hypothesis, denoted by H1 or

Ha, is the hypothesis that sample observations are influenced by some
non-random cause

8
Analyze
Hypothesis Testing

Example:

Suppose we wanted to determine whether a coin was fair and balanced. A null
hypothesis might be that half the flips would result in Heads and half, in Tails.
The alternative hypothesis might be that the number of Heads and Tails would
be very different.
Symbolically, these hypotheses would be expressed as
H0: P = 0.5
Ha: P ≠ 0.5
Suppose we flipped the coin 50 times, resulting in 40 Heads and 10 Tails. Given
this result, we would be inclined to reject the null hypothesis.

We would conclude, based on the evidence, that the coin was probably not fair
and balanced.

9
Analyze
Hypothesis Testing
Example:

Suppose a manager says his X employee’s performance level is 90%.

If the sample indicates his performance is 95% we can directly accept managers
statement. or
If our sample statistic says his performance is 46% then we can directly reject
manager’s statement.

Now, suppose our sample says his performance level is 88%, which is very close
to manager’s statement. We are not absolutely certain to accept or reject
manager’s statement.
What would be hypothesis statement?

H0: Performance level = 90%

Ha: Performance level < 90%

10
Analyze
Hypothesis Testing – Steps
● State the hypotheses: This involves stating the null and alternative
hypotheses. The hypotheses are stated in such a way that they are mutually
exclusive. That is, if one is true, the other must be false.

● Formulate an analysis plan: The analysis plan describes how to use sample
data to evaluate the null hypothesis. The evaluation often focuses around a
single test statistic.

● Analyze sample data: Find the value of the test statistic (mean, proportion,
etc.)

● Interpret results: Apply the decision rules. If the value of the test statistic is
unlikely, based on the null hypothesis, reject the null hypothesis.

11
Analyze
Hypothesis Testing – Decision Errors

Two types of errors can result from a hypothesis test:

● Type I error (Producer risk):

● A Type I error occurs when the researcher rejects a null hypothesis when it is true.
● The probability of committing a Type I error is called the significance level.
● This probability is also called alpha, and is often denoted by α.
● We will reach wrong conclusions and likely implement wrong solutions.

● Type II error (Consumer risk):

● A Type II error occurs when the researcher fails to reject a null hypothesis that is
false.
● The probability of committing a Type II error is called Beta, and is often denoted by β.
● The probability of not committing a Type II error is called the Power of the test.
● We will treat solution options as identical even though they aren’t.

12
Analyze
Hypothesis Testing – Confidence Interval

● A confidence interval is a range of values, calculated from a data set,

that gives us an assigned probability that the true value lies within that
range

● Usually, confidence intervals have an additive uncertainty:

Estimate ± margin of error

Ex: Saying that a 95% confidence interval for the mean is 35 ± 2,

means that we are 95% certain that the true mean of the population lies
somewhere at or between 33 to 37

13
Analyze
Hypothesis Testing – Confidence Interval
● The mean of a sample does not normally equate exactly to the mean
of the population from which the sample is taken

● An experimenter has more “confidence” that a sample mean, for

example, is close to the population mean whenever the sample size is
large, as opposed to whenever a sample size is small

● Statistical procedures quantify the uncertainty of a sample via a

confidence interval statement

e.g. mean ≤ 8.0 with 95% confidence

2.0 ≤ mean ≤ 8.0 with 90% confidence

Similar statements can be made about standard deviation and other

population characteristics

14
Analyze
Hypothesis Testing – Decision Rules
Two ways to describe decision rules:

● P- value:
● The strength of evidence in support of a null hypothesis is measured by the
P-value.

● The p-value is a numerical measure of the statistical significance of a

hypothesis test.

● If the data show that the null hypothesis has less than a 5% chance of being
right, we say it’s wrong.

● By convention, if the p-value is less than 0.05 (p < 0.05), we conclude that
the null hypothesis can be rejected.

15
Analyze
Hypothesis Testing – Decision Rules
Region of Region of Region of
Two ways to describe decision rules: rejection acceptance rejection

● Region of acceptance :
Critical 0 Critical
● The region of acceptance is a range of values. value value

● If the test statistic falls within the region of acceptance, the null hypothesis is
not rejected.

● The region of acceptance is defined so that the chance of making a Type I

error is equal to the significance level.

● The set of values outside the region of acceptance is called the region of
rejection.

● If the test statistic falls within the region of rejection, the null hypothesis is
rejected.

● In such cases, we say that the hypothesis has been rejected at the α level
of significance.

16
Analyze
Statistical Tools

Analysis
Analysis

Continuous
Continuous Discrete
Discrete
data
data data
data

One
OneSample
Sample Two
TwoSample
Sample 11--Proportion
Proportion 22--Proportion
Proportion
TTTest
Test TTTest
Test Test
Test Test
Test
Statistical parameter Between two DPU vs Target Between two binomial
vs Target statistical parameters Chi proportions
ChiSquare
Square
ANOVA
ANOVA Test
Test

To compare three or
To compare three or
more samples
more samples

17
Analyze
One Sample T Test
● Tells us whether or not a statistical parameter (average, standard
deviation, etc.) is different from a value of interest e.g. target

● Allows us to assess the “strength” of our conclusion (our probability of

being correct)

Example:

An automobile manufacturer has a target length for camshafts of 599.5 mm,

with an allowable range of ± 2.5 mm ( = 597.0 mm to 602.0 mm).
Data on the lengths of camshafts from supplier:
mean = 600.23 SD = 1.87
95% CI for mean is 599.86 to 600.60
The null hypothesis:
The camshaft from supplier are the same as the target value

18
Analyze
Two Sample T Test
● The two sample t is used to test whether or not the means of two
samples are the same

Example:

The same automobile manufacturer has data on another supplier and wants to
compare the two:
Supplier 1:
mean = 599.55 SD = 0.62
95% CI for mean is 599.43 to 599.67
Supplier 2:
mean = 600.23 SD = 1.87
95% CI for mean is 599.86 to 600.60

The null hypothesis:

The mean length of camshafts from supplier 1 is the same as the mean length of
camshafts from supplier 2

19
Analyze
ANOVA (Analysis Of Variance)
Purpose:

● To compare three or more samples to each other to see if any of the

sample means is statistically different from the others
● An ANOVA is used to analyze the relationships between several
categorical inputs (KPIVs) and one continuous output (KPOV)

When to use:

● In Analyze to confirm the impact of variables

● In Improve to help select the best option from several alternatives

20
Analyze
ANOVA (Analysis Of Variance)
● In the statistical world, inputs are sometimes referred to as factors

● The samples may be drawn from several different sources or under

several different circumstances. These are referred to as levels.

Example:

Compare on-time delivery performance of one product at three different

units (IDF1, 3 and 5).

“Unit” is considered to be a factor in the ANOVA and IDF1, 3 and 5 are

the “Levels”

21
Analyze
ANOVA (Analysis Of Variance)
● To tell whether the three or more options are statistically different,
ANOVA looks at three sources of variation:

● Total – Total variability among all observations

● Between – Variation between subgroup means (factor)
● Within – Random (chance) variation within each subgroup (noise, or
statistical error)

● In One-Way ANOVA, we look at how different levels of a single factor

affect a response variable

● In Two-Way ANOVA, we examine how different levels of two factors

and the interaction between those two factors affect a response
variable

22
Analyze
1 – Proportion Test
● 1 proportion test: When historic data (DPU) is available

Example:
BK500EI yari was running in cell2 from past 1 year with DPU of 8%. Later the
same model is moved to cell1 and they produced 400 units out of 210 units
passed. Calculate for statistical difference between these cell performance.

What would be hypothesis statement?

H0: DPU = 8%
Ha: DPU ≠ 8%
Test and CI for One Proportion
Test of p = 0.8 vs p not = 0.8 P value is less than .05.
Exact
Hence, Reject null
Sample X N Sample p 95% CI P-Value
hypothesis
1 190 400 0.475000 (0.425155, 0.525217) 0.000

23
Analyze
2 – Proportion Test
● 2 proportion test: Performs a test of two binomial proportions

Binomial: Situation where outcomes are mutually exclusive (e.g. Success or

Failure)

Example:
A corporation's purchasing manager need to authorize the purchase of twenty new
photocopy machines. After comparing many brands in terms of price, copy quality,
warranty, and features, he has narrowed the choice to two: Brand X and Brand Y. The
determining factor will be the reliability of the brands as defined by the proportion
requiring service within one year of purchase. Because corporation already uses both of
these brands, he was able to obtain information on the service history of 50 randomly
selected machines of each brand. Records indicate that six Brand X machines and eight
Brand Y machines needed service. With this information, he has to choose which brand
to purchase.

What would be hypothesis statement?

H0: P1 – P2 = P0
Ha: P1 – P2 ≠ P0

24
Analyze
2 – Proportion Test
Test and CI for Two Proportions
Sample X N Sample p
1 44 50 0.880000
2 42 50 0.840000
Difference = p (1) - p (2)
Estimate for difference: 0.04
95% CI for difference: (-0.0957903, 0.175790)
Test for difference = 0 (vs not = 0): Z = 0.58 P-Value = 0.564

P value is > .05.

Hence, there is not much evidence to reject null hypothesis

That is, the proportion of photocopy machines that needed service in the first
year did not differ depending on brand. As the purchasing manager, he need to
find a different criterion to guide the decision on which brand to purchase.

25