Lean Six Sigma Green Belt Cheat Sheet PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 18

GreyCampus

LEAN SIX SIGMA GREEN BELT


CHEAT SHEET

© www.greycampus.com
2

Includes formulas: what they are, when to use them, references

CONTENTS
LEAN CONCEPTS
• VSM Value Stream Mapping
• TAKT Time
• Batch Size
• SMED Singe Minute Exchange of Die
• Theory of Constraints
• TPM Total Productive Maintenance

SAMPLING
• Sample Size Calculator
• Single lot Sampling
• Dual lot Sampling
• Continuous Sampling
• Stratified Sampling
• Random Sampling

MSA
• MSA Measurement System Analysis
• Kappa MSA

DATA ANALYSIS
• Statistics Error Types
• Hypothesis Testing
• Pearson Correlation Co-efficient
• Central Limit Theorem
• FMEA Failure Mode and Effects Analysis

PROCESS CONTROL
• Attribute vs. Variable Data
• Control Charts
• VOC Voice Of the Customer
• Control Limits
• Process Capability
• Control plan

© www.greycampus.com
3

LEAN CONCEPTS
VSM Value Stream Mapping
• Value Stream Mapping is a tool used to understand a process and how much
value-added and non-value added time is spent on and between each activi-
ty.
• The VSM will include a data box of key statistics, such as:

Data Box Current State Future State % Improvement


Value Add Time
Non Value Add Time
Lead Time
People
Systems
System Touches
Inventory

TAKT Time
• Often referred to as the rate of customer demand.
• It is how often a product needs to be completed to meet customer demand.
• Formula = Effective Working Time / Average Customer Demand (for that time
period)
Batch Size

• To keep this explanation lean I’ll just write that moving to batch sizes of one
generally reduces cycle time and improves throughput.

SMED Singe Minute Exchange of Die

• SMED stands for Single Minute Exchange of Die, and covers the techniques
for obtaining a changeover time of less than 10 minutes (a single digit num-
ber of minutes).
• Basically, the SMED methodology consists of 6 steps:
◊◊ observe the current changeover process
◊◊ identify internal and external activities
◊◊ convert activities from internal to external setup
◊◊ increase efficiency of the remaining internal activities
◊◊ optimize the Startup time
◊◊ increase efficiency of external activities

Theory of Constraints
• The underlying assumption of Theory of Constraints is that organizations can
be measured and controlled by variations on three measures: Throughput,
Operating Expense, and Inventory.

© www.greycampus.com
4

• Throughput is money (or goal units) generated through sales. Operating Ex-
pense is money that goes into the system to ensure its operation on an on-
going basis. Inventory is money the system invests in order to sell its goods
and services.
• Theory of Constraints is based on the premise that the rate of goal achieve-
ment is limited by at least one constraining process. Only by increasing flow
through the constraint can overall throughput be increased. Assuming the
goal of the organization has been articulated (e.g.,“Make money now and in
the future”) the steps are:

1. IDENTIFY the constraint (the resource/policy that prevents the organiza-


tion from obtaining more of the goal)
2. Decide how to EXPLOIT the constraint (make sure the constraint’s time is
not wasted doing things that it should not do)
3. SUBORDINATE all other processes to above decision (align the whole sys-
tem/organization to support the decision made above)
4. ELEVATE the constraint (if required/possible, permanently increase capac-
ity of the constraint; “buy more”)
5. If, as a result of these steps, the constraint has moved, return to Step 1.
Don’t let inertia become the constraint.

TPM Total Productive Maintenance


• TPM is a program for planning and achieving minimal machine downtime
• Equipment and tools are literally put on “proactive” maintenance schedules
to keep them running efficiently and with greatly reduced downtime
• Machine operators take far greater responsibility for their machine upkeep
• Maintenance technicians are liberated from mundane, routine maintenance,
enabling them to focus on urgent repairs and proactive maintenance activi-
ties
• A solid TPM program allows you to plan your downtime and keep break-
downs to a minimum
• Without a strong TPM program, becoming truly Lean would be difficult or
impossible in an environment heavily dependent on machinery
• Buy-in at the shop floor level is generally quite high as TPM is an exciting
undertaking
• A robust TPM system consists of:

◊◊ Autonomous Maintenance
◊◊ Focused Improvement
◊◊ Education and Training
◊◊ Planned Maintenance
◊◊ Quality Maintenance
◊◊ Early Management and Initial Flow Control
◊◊ Safety, Hygiene and Pollution Control
◊◊ Administrative and Office TPM

© www.greycampus.com
5

• The metric used in Total Productive Maintenance environments is called OEE


or Overall Equipment Effectiveness
◊◊ OOE is measured as a percentage
◊◊ OOE = Availability * Performance * Quality
• Availability = % of scheduled production equipment is available for
production
• Performance = % number of parts produced out of best known pro-
duction rate
• Quality = % of good sellable parts out of total parts produced

SAMPLING
Sample Size Calculator

• To determine how large a sample you need to come to a conclusion about a


population at a given confidence level.
• Formula for creating a sample size to test a proportion:
◊◊ no = {Z2 pq}/{e2}
• no = required sample size
• Z = value from the Z table (found on-line or in stats books) for
confidence level desired
• p = estimated proportion of population that has the attribute
we are testing for
• q = 1-p
• e = precision {ie: if we want our proportion to be known within
10% then set ‘e’ at .05 and if we set the confidence interval at
95% and the sample gives a proportion of 43%, the true value at
95% confidence is between 38% and 48%}

• Formula for creating a sample size to test a mean:

• no = required sample size


• Z = value from the Z table (found on-line or in stats books) for
confidence level desired
• σ = variance of the attribute in the population
• e = precision in the same unit of measure as the variance
Single lot Sampling
• Single lot sampling is when your sample comes from a single lot. It is often
used in manufacturing when output from a single lot is sampled for testing.
• This may be used as a Lot Acceptance Sampling Plan (LASP) to determine
whether or not to accept the lot:

© www.greycampus.com
6

◊◊ Single sampling plans: One sample of items is selected at random from


a lot and the disposition of the lot is determined from the resulting
information. These plans are usually denoted as (n,c) plans for a sam-
ple size n, where the lot is rejected if there are more than c defectives.
These are the most common (and easiest) plans to use although not
the most efficient in terms `of average number of samples needed.
Dual lot Sampling
• Dual lot sampling is when your sample comes from 2 different but similar
lots. It is often used as part of an MSA or as part of hypothesis testing to de-
termine if there are differences in the lots.
Continuous Sampling
• Continuous sampling is used for the inspection of products that are not in
batches. The inspection is done on the production line itself, and each in-
spected item is tagged conforming or non-conforming. This procedure can
also be employed to a sequence of batches, rather than to a sequence of
items (known as Skip Lot Sampling).
Stratified Sampling
• Stratified Sampling is when the population is dived into non-overlapping
subgroups or strata and a random sample is taken from each subgroup. It is
often used in hypothesis testing to determine differences in subgroups.
Random Sampling
• Random sampling is a sampling technique where we select a group of sub-
jects (a sample) for study from a larger group (a population). Each individ-
ual is chosen entirely by chance and each member of the population has a
known, but possibly non-equal, chance of being included in the sample.

MSA
MSA Measurement System Analysis
• A Measurement System Analysis, abbreviated MSA, is a specially designed
experimentthat seeks to identify the components of variation in the mea-
surement.
• Since analysis of data is key to lean six sigma ensuring your data is accurate
is critical. That’s what MSA does – it tests the measurements used to collect
your data.
• Common tools and techniques of Measurement Systems Analysis include:
calibration studies, fixed effect ANOVA, components of variance, Attribute
Gage Study, Gage R&R, ANOVA Gage R&R, Destructive Testing Analysis and
others. The tool selected is usually determined by characteristics of the mea-
surement system itself.

◊◊ Gage R & R
• ANOVA Gauge R&R measures the amount of variability in-
duced in measurements that comes from the measurement

© www.greycampus.com
7

system itself and compares it to the total variability observed


to determine the viability of the measurement system.
• There are two important aspects on a Gauge R&R:

◊◊ Repeatability, Repeatability is the variation in measure-


ments taken by a single person or instrument on the same
item and under the same conditions.
◊◊ Reproducibility, the variability induced by the operators.
It is the variation induced when different operators (or
different laboratories) measure the same part.

• Formulas (this is best done using a tool such as Minitab or


JMP):

◊◊ yijk = μ + αi + βj + αβij + εijk


◊◊ Yijk = observation k with part i& operator j
◊◊ μ = population mean
◊◊ αi = adjustment for part i
◊◊ βj = adjustment for operator j
◊◊ αβij = adjustment for part i operator j interaction
◊◊ εijk = observation random ‘error’

◊◊ σ2y = σ2i + σ2j + σ2ij + σ2error


◊◊ σ2y = Variance of observation y
◊◊ σ2i = Variance due to part i
◊◊ σ2j = Variance due to operator j
◊◊ σ2ij = Variance due to part I operator j interaction
◊◊ σ2error = Variance of that observation due to ran-
dom ‘error’
• These formulas are used in the ANOVA Gage R &R to deter-
mine repeatability & reproducibility.

Kappa MSA

• MSA analysis for discrete or attribute data.


◊◊ Kappa (K) is defined as the proportion of agreement between raters
after agreement by chance has been removed.
◊◊ K = {Pobserved – Pchance}/{1 – Pchance}

• Pobserved = proportion of units classified in which the raters agreed


• Pchance = proportion of units for which one would expect agreement
bychance
◊◊ Generally a K < .7 indicates measurement system needs improvement
◊◊ K > .9 are considered excellent

© www.greycampus.com
8

DATA ANALYSIS
Statistics Error Types

• Type 1, Alpha or α errors


• Type I error, also known as an “error of the first kind”, an α error, or a
“falsepositive”: the error of rejecting a null hypothesis when it is actually
true. Plainly speaking, it occurs when we are observing a difference when
in truth there is none. An example of this would be if a test shows that a
woman is pregnant when in reality she is not. Type I error can be viewed
as the error of excessive credulity.
• Type 2, Beta or β errors
• Type II error, also known as an “error of the second kind”, a β error, or a
“false negative”: the error of failing to reject a null hypothesis when it is
in fact not true. In other words, this is the error of failing to observe a
difference when in truth there is one. An example of this would be if a
test shows that a woman is not pregnant when in reality she is. Type II
error can be viewed as the error of excessive skepticism.

Hypothesis Testing

• When to use what test: (The Six Sigma Memory Jogger II p 144)
• If comparing a group to a specific value use a 1-sample t-test (The Lean
Six Sigma Pocket Toolbook p 162)
◊◊ Tells us if a statistical parameter (average, standard deviation, etc.)
is different from a value of interest.
◊◊ Hypothesis takes the form Ho: µ = a target or known value
◊◊ This is best calculated using a template or software package. If
needed the formula can be found in the reference.
• If comparing 2 independent group averages use a 2-sample t-test (The
Lean Six Sigma Pocket Toolbook p 163)
◊◊ Used to determine if the means of 2 samples are the same.
◊◊ Hypothesis takes the form Ho: µ1 = µ2
• If comparing 2 group averages with matched data use Paired t-test
◊◊ The number of points in each data set must be the same, and they
must be organized in pairs, in which there is a definite relationship
between each pair of data points
◊◊ If the data were taken as random samples, you must use the inde-
pendent test even if the number of data points in each set is the
same
◊◊ Even if data are related in pairs, sometimes the paired t is still in-
appropriate
◊◊ Here’s a simple rule to determine if the paired t must not be used
- if a given data point in group one could be paired with any data
point in group two, you cannot use a paired t test

© www.greycampus.com
9

• If comparing multiple groups use ANOVA (The Lean Six Sigma Pocket
Toolbook p 173)
◊◊ Hypothesis takes the form Ho: µ1 = µ2 = µ3 = …

The smaller the p-value the more likely the groups are different.

Pearson Correlation Co-efficient

• In statistics, the Pearson product-moment correlation coefficient (sometimes


referred to as the PMCC, and typically denoted by r) is a measure of the cor-
relation (linear dependence) between two variables X and Y, giving a value
between +1 and −1 inclusive. It is widely used in the sciences as a measure of
the strength of linear dependence between two variables.
◊◊ Remember Pearson measures correlation not causation.

• A value of 1 implies that a linear equation describes the relationship between


X and Y perfectly, with all data points lying on a line for which Y increases as
X increases. A value of −1 implies that all data points lie on a line for which Y
decreases as X increases. A value of 0 implies that there is no linear relation-
ship between the variables.

• The statistic is defined as the sum of the products of the standard scores of
the two measures divided by the degrees of freedom Based on a sample of
paired data (Xi, Yi), the sample Pearson correlation coefficient can be calcu-
lated as:

◊◊ r = {1/(n-1)}∑ [(Xi – X(avg))/(Sx)][(Yi-Y(avg))/(Sy)]


◊◊ where
• n = sample size
• Xi = the value of observation I in the X plane
• X(avg) = the average X value
• Sx = the standard deviation of X
• Yi = the value of observation i in the Y plane
• Y(avg) = the average Y value
• Sy = the standard deviation of Y

Central Limit Theorem


• In probability theory, the central limit theorem (CLT) states conditions under
which the sum of a sufficiently large number of independent random vari-
ables, each with finite mean and variance, will be approximately normally
distributed.
• Let X1, X2, X3, ..., Xn be a sequence of n independent and identically dis-
tributed(i.i.d) random variables each having finite values of expectation
µ and variance σ2 > 0.

© www.greycampus.com
10

• The central limit theorem states that as the sample size n increases, the-
distribution of the sample average of these random variables approaches
the normal distribution with a mean µ and variance σ2 / n irrespective of
the shape of the original distribution.
• By using the central limit theorem we can apply tools that require a
normal distribution even when the distribution of the population is
non-normal. However, be careful when interpretingresults if you use the
CLT to analyze averages rather than samples from the direct population.
Remember your analysis is based on averages which can be dangerous.
In many cases it’s safer to find a tool that doesn’t require normally dis-
tributed data when analyzing non-normal data.
• • Applying CLT to data analysis:
◊◊ Take 10 data points from the distribution & average them. Then
take 29 other samples of 10 data points averaged. Then use these 30
data averaged data points to do your analysis. This converts your
data from its original distribution to a normal distribution.
◊◊ As long as n (sample size) is large enough, the sampling distri-
bution will be normal and the mean will be representative of the
population mean μ.
◊◊ No matter what the parent looks like the child will look rea-
sonably normal by 30!
◊◊ Formulas:
• Sampling distribution for the mean has SD = σ / √n
• For variables data, applying the 95% rule we expect μ to lie in
the interval:
• 95% Confidence Interval for the Population Mean =
• Sample size: h (maximum error allowed) =
• Attribute data:
• 95% Confidence Interval for the Population Pro-
portion =
• Sample size: ( )

FMEA Failure Mode and Effects Analysis

• This tool helps determine where to focus improvement efforts by analyzing


severity offailures, probability of occurrence of an error, and likelihood of de-
tection of an error.

© www.greycampus.com
11

• An RPN (Risk Priority Number) is computed by multiplying these factors to-


gether.Processes/Steps with the highest RPN should be addressed first.

PROCESS CONTROL
Attribute vs. Variable Data
• Different Tools are best able to handle different types of date (for example
control charts).
• For six sigma data analysis Discrete or Attribute data is considered the same
thing. Discrete data is any data not quantified on an infinitely divisible nu-
merical scale. Discrete data has boundaries and includes any number that
must be an integer. Discrete examples include day of the week; hour of the
day; age in years; counts; income; an non-numeric translated to a numeric
value (i.e.: good/bad, on-time/not on-time, pass/fail, primary colors, eye col-
or, grade, method).
• Variable or Continuous data is any data on a continuous scale. Examples in-
clude length,diameter, temperature, weight, time.
Control Charts
• A Control Chart is simply a run chart with statistically based limits.
• Within the subject of control charts Voice of the Customer (VOC) is the cus-
tomer requiredspecifications or the Upper Spec Limit (USL) and Lower Spec
Limit (LSL).
• The Voice of the Process is the Upper Control Limit (UCL), and Lower Control
Limit (LCL). It is 3 standard deviations from the process mean and is what the
process will deliver (99.7% of the time).
• When to use what chart:
◊◊ Variable Individual data: use XmR or I or mR (use moving range to mea-
sure control)
• XmR or I
◊◊ Called XmR since we use the Moving Range relative to X
◊◊ Also called I since it is based on Individual data
◊◊ This chart will work for virtually any situation
◊◊ K = # of subgroups
◊◊ Avg(X) = ∑X/k
◊◊ Rm = | (Xi+1 – Xi) |
◊◊ Avg (Rm) = ∑R/(k-1)
◊◊ E2 is based on sample size & is in a table in the reference
◊◊ UCL = avg(X) +E2 * Avg (Rm)
◊◊ LCL = avg(X) +E2 * Avg (Rm)

© www.greycampus.com
12

• mR
◊◊ K = # of subgroups
◊◊ Avg(X) = ∑X/k
◊◊ Rm = | (Xi+1 – Xi) |
◊◊ Avg (Rm) = ∑R/(k-1)
◊◊ D3 & D4 are based on sample size & is in a table in the refer-
ence
◊◊ UCL = D4 * Avg (Rm)
◊◊ LCL = D3 * Avg (Rm)
◊◊ Variable data of group sizes 2-10: use Xbar & R (use range to measure con-
trol)
• Xbar
◊◊ K = # of subgroups
◊◊ Avg(X) = Avg of X for subgroup k
◊◊ Avg(Avg(X)) = ∑Avg(X)/k
◊◊ Avg (R) = ∑R/k
◊◊ A2 is based on sample size & is in a table in the reference
◊◊ UCL = Avg(Avg(X)) + A2 * Avg(R)
◊◊ LCL = Avg(Avg(X)) – A2 * Avg(R)
• R
◊◊ K = # of subgroups
◊◊ Avg (R) = ∑R/k
◊◊ D4 & D3 are based on sample size & is in a table in the refer-
ence
◊◊ UCL = D4 * Avg (R)
◊◊ LCL = D3 * Avg (R)
◊◊ Variable data of group sizes 10 or more: use Xbar & S (use standard devi-
ation to measure control)
◊◊ Xbar
◊◊ K = # of subgroups
◊◊ Avg(X) = Avg of X for subgroup k
◊◊ Avg(Avg(X)) = ∑Avg(X)/k
◊◊ Avg (s) = ∑s/k
◊◊ A3 is based on sample size & is in a table in the reference
◊◊ UCL = Avg(Avg(X)) + A3 * Avg(s)
◊◊ LCL = Avg(Avg(X)) – A3 * Avg(s)

© www.greycampus.com
13

◊◊ S
◊◊ K = # of subgroups
◊◊ Avg (s) = ∑s/k
◊◊ B3 & B4 are based on sample size & is in a table in the
refernce
◊◊ UCL = B4 * Avg(s)
◊◊ LCL = B3 * Avg(s)
• Note: Defect data is a failure to meet one of the acceptance criteria;
Defective data is when an entire unit fails to meet acceptance criteria. A
defective unit may have multiple defects.
• Attribute Defect Data Constant sample size: use C
◊◊ C
◊◊ c = number of defects for each subgroup
◊◊ k = # of subgroups
◊◊ avg (c) = ∑c/k
◊◊ UCL = avg (c) + 3√avg (c)
◊◊ LCL = avg (c) - 3√avg (c)
• Attribute Defect Data Variable sample size : use U
• U
• u = c/n for each subgroup
• avg (u) = ∑c/∑n
• UCL = avg (u) + 3√(avg (u)/n)
• LCL = avg (u) + 3√(avg (u)/n)
• Attribute Defective Data Constant sample size: use nP for number defec-
tive
• nP
◊◊ most reliable when sample size greater than 50
◊◊ np = # defective units for each subgroup
◊◊ k = # of subgroups
◊◊ avg (np) = ∑np/k
◊◊ UCL = avg (np) + 3√[avg(np) * {1- avg(p)}]
◊◊ LCL = avg (np) - 3√[avg(np) * {1- avg(p)}]
◊◊ Attribute Defective Data Variable sample size: use p for fraction defective
• P
◊◊ most reliable when sample size greater than 50
◊◊ p = np/n = % defective units for each subgroup

© www.greycampus.com
14

◊◊ § avg(p) = ∑np/∑n
◊◊ UCL = avg (p) + 3√([avg(np) * {1- avg(p)}]/n)
◊◊ LCL = avg (p) - 3√([avg(np) * {1- avg(p)}]/n)
◊◊ Process is known to be in control & want to examine small shifts: use
CUSUM or EWMA
• CUSUM (CUmulative SUM chart)
◊◊ The Cusum chart can detect process shifts more rapidly than
the XmR or XbarR. If a trend develops, it’s evidence that the
process has shifted.
◊◊ To compute:
• n = number of data points
• Avg(x) = ∑X/n
• S(i) = Cumulative Sum of the difference between the
current value and the average.
• The cumulative sum is not the cumulative sum of the-
values. Instead it is the cumulative sum of differences-
between the values and the average. Because the aver-
age is subtracted from each value, the cumulative sum
also ends at zero.
• If S(i) is positive or negative for more than a few points
in a row it is an indication there may be a change in
the process.
• EWMA (Exponentially Weighted Moving Average)
◊◊ The Exponentially Weighted Moving Average (EWMA) is a
statistic for monitoring the process that averages the data
in a way that gives less and less weight to data as they are
further removed in time.
◊◊ EWMAt = λYt + ( 1-λ ) EWMAt-1 for t = 1, 2, ..., n.
• EWMA0 is the mean of historical data (target)
• Yt is the observation at time t
• n is the number of observations to be monitored in-
cluding EWMA0
• 0 < λ < 1 is a constant that determines the depth of
memory of the EWMA.
• The parameter λ determines the rate at which ‘older’
data enter into the calculation of the EWMA statistic.
A value of λ = 1 implies that only the most recent mea-
surement influences the EWMA (degrades to Shewhart
chart). Thus, a large value of λ = 1 gives more weight to
recent data and less weight to older data; a small val-
ue of λ gives more weight to older data. The value of is
usually set between 0.2 and 0.3 (Hunter) although this
choice is somewhat arbitrary.

© www.greycampus.com
15

• s2 ewma = ( λ/(2- )) s2
• The center line for the control chart is the target value
or EWMA0. The control limits are:
◊◊ UCL = EWMA0 + ksewma
• • LCL = EWMA0 - ksewma
• where the factor k is either set equal 3 or chosen using
the Lucas and Saccucci (1990) tables. The data are as-
sumed to be independent and these tables also assume
a normal population.

• Out of Control Conditions:


◊◊ If 1 or more points falls outside of the UCL (Upper Control Limit) or LCL
(Lower Control Limit). (i.e.: any point more than 3 standard deviation
from the mean).
◊◊ If 2 of 3 successive points fall beyond 2 standard deviations from the
mean.
◊◊ If 4 of 5 successive points fall beyond 1 standard deviations from the
mean.
◊◊ If there is a run of 6 or more points that are either successively higher or
successively lower.
◊◊ If 8 or more successive points fall on either side of the mean.
◊◊ If 15 points in a row fall within 1 standard deviation from the mean.(Ex-
amine to determine if the process has become better controlled).
VOC Voice Of the Customer
• Every lean Six Sigma project needs to understand the Voice of the Customer.
You can not truly exit define phase without an understanding of what’s im-
portant to the customer.
• In Lean only process that add value to the customer are value-add processes.
You need to understand the customer (i.e.: what would the end-customer pay
for) to understand if a step is technically a value-added step in a Lean analy-
sis.
• In control charts the VOC is translated to Upper Spec Limit (USL) & Lower
Spec Limit (LSL) (see control charts above).
Control Limits

• See control charts above. Control Limits are the ‘voice of the process’ and
based on the
• standard deviation and variability of the process.

Process Capability (The Black Belt Memory Jogger p 95)


• Measures the capability of a process to meet customer specifications.
• The higher the Cp, Cpk, Pp, or Ppk, the better, the less variance there is in
the process.

© www.greycampus.com
• Short-Term Capability indices:
◊◊ Used if long-term data is not available or if process has changed recently.
◊◊ Uses short term process variation to determine process capability
◊◊ Cp tells you if the process is capable: If Cp > 1 then the process is capable
(the process spread fits within tolerance). While Cpk also tells you if the
process is centered; if Cpk < 1 while Cp > 1 then the process is not centered
and should first be centered.
◊◊ Cp
• The ratio of customer-specified tolerance to 6 standard deviations
of the short-term process variation.
• Gives an indication of what the process could perform to if the
mean of the data was centred between the specification limits.
• Sometimes referred to as process potential
• Formula: Cp = USL - LSL / 6sMR
◊◊ Cpk
• The ratio of the distance between the process average and the
closest specification limit to 3 standard deviations of short term
process variation.
• More realistic measure of process than Cp because it uses data
mean.
• Formula: Cpk = min{USL- x/3sMR, x- LSL/3sMR}
• Long Term Capability indices:
◊◊ Used if long-term data available & representative of the current process.
◊◊ Uses long-term process standard variation
◊◊ Pp
• The ratio of customer-specified tolerance to 6 standard deviations
of the long-term process variation.
• Gives an indication of what the process could perform to if the
mean of the data was centred between the specification limits.
• Sometimes referred to as process potential
• Formula: Pp = USL - LSL / 6s
◊◊ Ppk
• The ratio of the distance between the process average and the clos-
est specification limit to 3 standard deviations of long term process
variation.
• More realistic measure of process than Pp because it uses data
mean.
• Formula: Ppk = min { USL - x /3s, x - LSL/3s }
Process Capability (The Black Belt Memory Jogger p 95)

• The control plan is the plan used to control the process. Within six sigma it
is used in the control phase & after project closure to ensure project improve-
ments are sustained.
• There a various control plan templates, but what needs to be captured is
what is being measured; how it is used, what sampling method is used, who
owns the control chart, where it is located, and what conditions constitute
loss of control or improvement that require corrective action.
• Usually Cp, Cpk, Pp, and Ppk (see process capability above) are measured as
part of the control plan, and often process control charts (see control charts
above) are the key control chart used.
ABOUT GREYCAMPUS
GreyCampus is a leading provider of on-demand training that address the unique learning needs of
professionals, delivered as online self-learning, live online training or in-person classroom training.
Our aim is to provide quality training enabling professionals to achieve their certification and
career enhancement goals. We offer training for certifications in areas of Big Data & Hadoop, Project
Management, IT Service Management, Quality Management, Python Programming, Agile Training
Coaching & Certification and Workplace Tools.

TRAINED OVER 30,000 REACH ACROSS EXAM PASS RATE OF COURSES ACCREDITED BY
PROFESSIONALS 50+ COUNTRIES OVER 97 % LEADING GLOBAL BODIES

ACCREDITATIONS & ASSOCIATIONS

Provider ID : 3871

DISCLAIMER
“PMI®”, “PMBOK®”, “PMP®” “CAPM®” and “PMI-ACP®” are registered marks of the Project Management Institute, Inc.

The Swirl logo™ is a trade mark of AXELOS Limited.


ITIL® is a registered trade mark of AXELOS Limited.
PRINCE2® is a Registered Trade Mark of AXELOS Limited.

IASSC® is a registered mark of International Association for Six Sigma Certification.

© www.greycampus.com

You might also like