Chapter Two

Statistical Tools For Problem Solving

Daniel Ab.
Bahir Dar Institute of Technology (BiT)
Faculty of Mechanical and Industrial Engineering
2.1 Statistical Thinking
Statistical Thinking is a philosophy of learning and acting based on
the following fundamental principles:
• All work occurs in a system of interconnected processes,
• Variation exists in all processes, and
• Understanding and reducing variation are keys to success.

2
Key concepts of statistical thinking
1. Process and Systems thinking
2. Variation
3. Analysis to increase knowledge
4. Taking action
5. Improvement
Role of data in statistical thinking
• Quantify variation
• Measure effects

Statistical thinking without a process view

• People have problems understanding the problem and their role in its
solution
• It is difficult to define the scope of the problem
• It is difficult to get root causes
• People get blamed when the process is the problem
• Process management is ineffective
3 • Improvement is slowed
“Youcan’t improve a processthat you don’tunderstand”
Statistical thinking without understanding variation
• There will be management by the last data point
• There’s lots of fire fighting
 Using special cause methods to solve common cause problems
• Understanding the process is handicapped
 Learning is slowed
• Process management is ineffective
• Improvement is slowed

Statistical thinking without data

• Everyone is an expert
• Historical memory is poor
• Difficult to get agreement on:
 What the problem is
 What success looks like
 Progress made
4 • Process management is ineffective
• Improvement is slowed
Use of statistical Thinking

Depends on levels of activity and job responsibility. The following diagram

shows the status which we use statistical thinking, where to apply and who to
use it.
Level of the problem where to apply who to use it

Where we are headed Strategic Executives

Managerial Process to Managerial Managers

guide us

Where the work gets done Operational Workers

2.2 Measurement and Data Collection
If we adopt the definition of quality as ‘meeting the customer
requirements’, there is already the need to consider the quality of
design and the quality of conformance to design.

To achieve quality therefore requires:

● An appropriate design;
● Suitable resources and facilities (equipment, premises, cash, etc.);
● The correct materials;
● People, with their skills, knowledge and training;
● An appropriate process;
● Sets of instructions;
● Measures for feedback and control.
All are data which need to be measured and collected properly
Data Collection
● Data should form the basis for analysis, decision and action, and their form and
presentation will obviously differ from process to process.
● Information is collected to discover the actual situation. It may be used as a
part of a product or process control system and it is important to know at the
outset what the data are to be used for.
● For example, if a problem occurs in the amount of impurity present in a
product that is manufactured continuously, it is not sufficient to take only one
sample per day to find out the variations between – say – different operator
shifts.
● It is not unusual to find that decisions and action are based on misconceived
or biased data. In other words, full consideration must be given to the reasons
for collecting data, the correct sampling techniques.
● The methods of collecting data and the amount collected must take account
of the need for information and not the ease of collection; there should not be a
disproportionate amount of a certain kind of data simply because it can be
collected easily.
Types of Data
Numeric information will arise from both counting and measurement.
● Data that arise from counting can occur only in discrete steps. E.g. there can be
only 0, 1, 2, etc., defectives in a sample of 10 items, there cannot be 2.68
defectives.
● The number of defects in a length of cloth, the number of typing errors on a
page, the presence or absence of a member of staff, are all called attributes. As
there is only a two-way or binary classification, attributes give rise to discrete
data, which necessarily varies in steps.
● Data that arise from measurements can occur anywhere at all on a continuous
scale and are called variable data. The weight of a tablet, share prices, time
taken for a rail journey, age, efficiency, and most physical dimensions, are all
variables, the measurement of which produces continuous data.
● If variable data were truly continuous, they could take any value within a given
range without restriction. However, owing to the limitations of measurement,
all data vary in small jumps, the size of which is determined by the instruments
in use.
● The statistical principles involved in the analysis of whole numbers are not
usually the same as those involved in continuous measurement.
 …..cont’d
Generally, we can classified in two basic categories:
1. Categorical data – Data for discreet variables
2. Continuous data – Data for continuous variables

1. Categoricalvariables:
● Numbers and proportions in each category. e.g. 5 points Likert scale – Highly
satisfied, Satisfied, Neither satisfied nor dissatisfied, dissatisfied, highly
dissatisfied)
2. Continuous variables
1. Distributions
❑ Normal (Gaussian)
❑ Non-parametric
2. Centraltendency
❑ Mean
❑ Median
3. Scatter
❑ Standard deviation
❑ Range
❑ Inter-quartile range
❑ Standard error of the mean
Accuracy and Precision
Accuracy
● Accuracy is how close you are to the true value. The true value is sometimes
called thenominal
theoretical
valuevalue that would be obtained by a perfect measurement.
● The accuracy of a data set or a measuring instrument refers to the degree
of uniformity of the observations around a desired value such that, on
average, the target value is realized.

● Error - the result of a measurement minus a true value of the measure.

● Systematic Error (Bias)- the mean that would result from an infinite number
of measurements of the same measure and carried out under repeatability
conditions, minus a true value of the measure and.
● These errors result from improper condition and it affects all measurements
in the same way i.e. the accuracy of the measurement.
Example: Ascale that reads 0.5 pounds with nothing on it.
Aclock that is 5 minutes slow.
● Systematic errors can therefore be avoided, i.e., they are determinate errors.
 …..cont’d
● Random Error - the result of a measurement minus the mean that would
result from an infinite number of measurements of the same measure and
carried out under repeatability conditions.
● They are caused by unknown and unpredictable changes in the measuring
instruments or in the environmental conditions which affect the precision
of the measurement.
● Such errors are inherent in the measuring system, and are unavoidable, i.e.,
they are indeterminate errors.
Example: Reading a measuring device wrong.
 …..cont’d
Precision
● The precision of a data set or a measuring instrument refers to the degree
of variability of the observations. Observations may be off the target value
but still considered as precise.
● Precision is how close two or more measurements are to each other, i.e.
consistency of individual measurement.
● If you are precise, that doesn’t necessarily mean you are accurate. However,
if you are consistently accurate, you are also precise.
Example: Target value = 5.25mm

5.25mm
5.25mm

x x x
x x x
x x x x x x x x x
x x x x x x x x x
x x x x x 5.25mm x x x x x x
x x x x x x x x x x x x x
x x x x x x x x x x x x x x x

Accurate observation precise observation Accurate and precise observation

…..cont’d
2.3 Populations and Samples-Sampling Issues
 Sampling is the process of obtaining samples.
 A sampling design is a description of the procedure by which the
observations in a sample are to be chosen. It does not necessarily deal
with the measuring instrument to be used.
Example: every tenth item produced
 Population: It is the entire group of elements from which data will be
collected. It includes all members that meet a set of specifications or a
specified criterion.
 A Sampling unit: It is an individual element or a collection of non-
overlapping elements from the population that make up the frame.
 A Sampling frame: It is a list of all sampling units within the
population who can be sampled.
 Element: It is an object (or group of objects) for which data or information
is gathered. It is a single member of any given population.
 Sample Space: It is a set that consists of all possible outcomes of a
randomexperiment.
Sampling Design and Schemes
• A major objective of any sampling design or scheme is to select the sample in
such a way as to accurately portray the population from which it is drawn. After
all, a sample is supposed to be representative of the population.
• Sampling in general has certain advantages, if the measurement requires
destroying the item being measured (destructive testing).
• If we draw an object from urn, we have the choice of replacing or not replacing
the object into urn before we draw again. In the first case a particular object can
come up again and again, whereas in the second it can come up only once.
● Sampling with replacement - Sampling where each member of a
population may be chosen more than once.
● Sampling without replacement – Sampling where each member
cannot be chosen more than once.
• A finite population that is sampled with replacement can theoretically be
considered infinite since samples of any size can be drawn without exhausting
the population. For most practical purposes, sampling from a finite population
that is very large can be considered as sampling from as infinite population.
Sampling Design Process

judge
2.4 Fundamental Statistical Measures
Measures of Central Tendency
Ameasure of central tendency gives a single value that acts as a representative or average
of the values of all the outcomes of your experiment.The main measure of central
tendency we will use is the arithmetic mean.While the mean is used the most, two
other measures of central tendency are also employed.These are the median and the
mode. Each has advantages and disadvantages, depending on the data and the intended
purpose.

Mean – if we are given a set of n numbers of values, the mean is the average of these
n values.
Median – is that value x for which P(X<x)<=1/2 and P(X>x)<=1/2. In other
words, the median is the value where half of the values of the total sample are
larger than the median, and half of them are smaller than the median.
Example: Consider the following set of integers:
S = {1, 6, 3, 8, 2, 4, 9}
If we want to find the median, we need to find the value, x, where half the values are
above xand half the values are below x. begin by ordering the list in ascending order.
S= {1, 2, 3, 4, 6, 8, 9}
4 is at the middle wayof the list and it is the median.
If the number of values are even number, we will find two values at the middle
way of the list and the median can be any number between these two numbers
but, commonly it is taken the average value of these numbers:
Example: S= {1, 2,3, 4, 6, 8, 9, 12}
4 and 6 are at the middle wayof the list and the median is 5 (the average)

Mode – is the vale that occurs most often.

Example: consider the following rolls of a ten-sided die:
R = {2, 8, 1, 9, 5, 2, 7, 2, 7, 9, 4, 7, 1, 5, 2}
The number that appears the most is 2. therefore, the mode is 2.
If there are more than one values which appear the most:, the distribution is
called,
● Bimodal– if there are two values
● Tri-modal– if there are three values
● Multimodal– if there are more than three values
Measures of Variation (Dispersion)

Consider the following two sets of integers:

S={5, 5, 5, 5, 5,5} and R={0,0,0, 10, 10, 10}

Here, the mean is 5 for both S and R. but they are two vastly different data sets.
We need another descriptive statistic besides a measure of central tendency,
which we shall call a measure of variationor measureof dispersion.

We can measure the dispersion or scatter of the values of our data set about the
mean of the data set. If the values tend to be concentrated near the mean, then
this measure shall be small, while if the values of the data set tend to be
distributed far from the mean, then the measure will be large.
The three measures of variations that are usually used are called the range,
varianceand standard deviation.
Variance – is donated by σ2
For a set of n numbers x1, x2, …, xn and if μ is the mean of the values, the variance
is given by:
[(X1 - μ)2+(X2 - μ)2+…+(Xn - μ)2]
σ2 =
n
Therefore, the variance is a nonnegative number.

Standard Deviation: The positive square root of the variance.

Example:T= {75, 80, 82, 87, 96}
μ =84, σ2 = 50.8 and standard deviation = 7.1274118

Range: It is the difference between the maximum and minimum values of a

measured parameter.
2.5 Seven Quality Control Tools for Data Collection and Analysis
• The purpose of statistical quality control is to ensure, in a cost efficient
manner, that the product shipped to customers meets their specifications.
• Inspecting every product is costly and inefficient, but the consequences
of shipping non conforming product can be significant in terms of
customer dissatisfaction.
• Statistical Quality Control is the process of inspecting enough product from
given tools to ensure a specified quality level.
• To make rational decisions using data obtained on the product, or process,
or from the consumer, organizations use certain graphical tools.There are
many ways to implement process control. Key monitoring and investigating
tools include:
1. Check Sheet tally sheet
2. Histogram
3. Flow Chart procces chart
4. Pareto Diagram
5. Cause and Effect Diagram fish bon
6. Scatter Plot Diagram
7. Control Chart
Check Sheet (Tally Sheet)
• Concentrating defects into major causes or categories.
• Facilitate systematic record keeping or data collection.
Observations are recorded as they happen, which reveals
patterns or trends.
• A check sheet is a structured, prepared form for collecting and
analyzing data.This is a generic tool that can be adapted for a
wide variety of purposes.
• It is a collection of daily/weekly or monthly records (based on
the frequency of the occurrence of the problem or cause).
Product No.: Product Name:
Date: Shift: Operators Name:

Problems/Causes of Days/Weeks/Months
problems 1 2 … m Total
Problem 1 / Cause 1 No. of Occurrences |||| … ||
Problem 2 / Cause 2 ||| |||||
… |||||| |
Problem n / Cause n || ||| Occurrence

Total number Inspected

Total number rejected
Rejection rate (%)

Figure: Structure of Check Sheets

After recording the data like this, the team will discuss for each occurrences
and their causes in detail. After thorough discussion on the data, solutions for
the problems and how to sustain improvements are proposed.

Phases of Using Check Sheet

It has two phases, namely:
1.Define – specify from which section of the company to collect data,
what problems to address, and the time period.
2.Measure – record the occurrences of each problem in each specific period.

Guidelines to use check sheet

● Use the Check Sheet when data can be observed and collected
repeatedly by the same person or at the same location.
● Use the Check Sheet when collecting data on the frequency or
patterns of events, problems, defects, defect location, defect causes,
etc.
Histogram bar graph

• The data obtained from the check sheets can be put into a histogram.
A histogram is a snapshot of the variation of a product or the results
of a process. It often forms the bell-shaped curve which is
characteristic of a normal process.
• This tool is a bar graph which displays a frequency distribution of
the occurrence of the various measurements. It gives quick look at
the trends of a process.
• Histogram helps to depict and analyze the central tendency or mean
of the data, and its variation or spread. It also shows the range of
measurements, which defines the process capability, whether the
data is falling inside the bell-shaped curve and within specifications.
• The variable being measured is along the horizontal x-axis, and is
grouped into a range of measurements. The frequency of occurrence
of each measurement is charted along the vertical y-axis.
A histogram can show characteristics of the process being measured, such
as:
• Do the results show a normal distribution, a bell curve? If not, why
not?
• Does the range of the data indicate that the process is capable of
producing what is required by the customer or the specifications?
• How much improvement is necessary to meet specifications? Is this
level of improvement possible in the current process?

negative skewed
Flow Chart
• It is picturing the process.
• It uses to allow a team to identify the actual flow or sequence of events in a
process that any product or service follows.
Phases
Define– this phase is identifying which process to study and draw the process
flow of the current process.
Measure– identifying problems in the current flow chart
Improve– Improving the flow chart to solve the identified problem

To draw a flow chart, you need to know:

 The Process
 Each activities in the process
 The precedence of the activities in the process
 The characteristics of the activities
 The decisions to be made, the data used and types of documents and
activities in the process.
Purpose:
Visual illustration of the sequence of operations required to complete a
task:
 Schematic drawing of the process to measure or improve.

 Starting point for process improvement

 Potential weakness in the process are made visual.

 Picture of process as it should be.

Benefits:
 Identify process improvements

 Understand the process

 Shows duplicated effort and other non-value-added steps

 Clarify working relationships between people and organizations

 Target specific steps in the process for improvement.

Symbols Used to Draw Flow Chart
 SetTool

Purge System

Position material for

press operation

Mold Part

Remove Part
Light
Visual
dirt Bad
Buff part Inspection by Scrap
near dirt
operator
surface
Part OK
Trim Part

Assemble part

Functional
Part no
Inspection by Scrap
good
operator
Part OK
Package
Pareto Diagrams

 Helps prioritize problems by arranging them in decreasing order of

importance. In an environment of limited resource, these diagrams help
companies decide on the order in which they should address problems.
 It is to focus efforts on the problems that offers the greatest potential for
improvement by showing their relative frequency or size in decreasing bar
graph.
 Pareto analysis helps graphically display results so the significant few
problems emerge from the general background. It tells you what to work
on first (Separating ‘Vital Few’ from ‘Trivial Many’).
 This diagram is also called the ‘80/20 rule’ and shows itself in many
ways. For example:
• 80% of sales are generated by 20% of customers.
• 80% of Quality costs are caused by 20% of the problems.
• 20% of stock lines will account for 80% of the value of the stock.
As a basic Quality Improvement tool, Pareto Analysis can:
 define categories of defects which cause a particular output
(product, service, unit) to be defective;
 count the frequency of occurrence of each defect;
 display graphically as a bar chart, sorted in descending order, by
frequency of defect;
 use a second y axis to show the cumulative % of defects .

How to use it:

1. Gather facts about the problem, using Check Sheets or
Brainstorming, depending on the availability of information.
2. Rank the contributions to the problem in order of frequency.
3. Draw the value (errors, facts, etc) as a bar chart.
4. It can also be helpful to add a line showing the cumulative
percentage of errors as each category is added. This helps to
identify the categories contributing to 80% of the problem.
5. Review the chart – if an 80/20 combination is not obvious, you
may need to redefine your classifications and go back to Stage 1
or 2.
Cause-and-Effect Diagram (Fish Bone Diagram)
• A Cause-and-Effect Diagram is a tool that helps identify, sort, and
display possible causes of a specific problem or quality characteristic.
• It graphically illustrates the relationship between a given outcome
and all the factors that influence the outcome. This type of diagram
is sometimes called an "Ishikawa diagram” because it was invented
by Kaoru Ishikawa, or a "fishbone diagram" because of the way it
looks.
When should a team use a Cause-and-EffectDiagram?
Constructing a Cause-and-Effect Diagram can help your team when you
need to
 Identify the possible root causes, the basic reasons, for a specific
effect, problem, or condition.
 Sort out and relate some of the interactions among the factors
affecting a particular process or effect.
 Analyze existing problems so that corrective action can be taken.
Basic Layout of Cause-and-Effect Diagram
(Fish Bone Diagram)
Why should we use a Cause-and-Effect Diagram?

● Helps determine the root causes of a problem or quality

characteristic using a structured approach.
● Encourages group participation and utilizes group knowledge
of the process.
● Uses an orderly, easy-to-read format to diagram cause-and-
effect relationships.
● Indicates possible causes of variation in a process.
● Increases knowledge of the process by helping everyone to
learn more about the factors at work and how they relate.
● Identifies areas where data should be collected for further
study.
How do we develop a Cause-and-Effect Diagram?
• When you develop a Cause-and-Effect Diagram, you are constructing a
structured, pictorial display of a list of causes organized to show their
relationship to a specific effect.
• Notice that the diagram has a cause side and an effect side. The steps for
constructing and analyzing a Cause-and-Effect Diagram are outlined
below.
Step 1 - Identify and clearly define the outcome or EFFECT to be
analyzed
● Decide on the effect to be examined. Effects are stated as particular quality
characteristics, problems resulting from work, planning objectives, and the
like.
● Use Operational Definitions. Develop an Operational Definition of the
effect to ensure that it is clearly understood.
● Remember, an effect may be positive (an objective) or negative (a
problem), depending upon the issue that’s being discussed.
➢ Using a positive effect which focuses on a desired outcome tends to
foster pride and ownership over productive areas. This may lead to an upbeat
atmosphere that encourages the participation of the group. When possible,
it is preferable to phrase the effect in positive terms.

➢ Focusing on a negative effect can sidetrack the team into justifying

why the problem occurred and placing blame. However,it is sometimeseasier for a
team to focus on what causes a problem than what causes an excellent
outcome. While you should be cautious about the fallout that can result
from focusing on a negative effect, getting a team to concentrate on things
that can go wrong may foster a more relaxed atmosphere and sometimes
enhances group participation.
Step 2 - Using a chart-pack positioned so that everyone can see it, draw
the SPINE and create the EFFECT box.
● Draw a horizontal arrow pointing to the right.This is the spine.
● To the right of the arrow, write a brief description of the effect or outcome which
results from the process.
● Draw a box around the description of the effect.
Step 3 - Identify the main CAUSES contributing to the effect being studied.
These are the labels for the major branches of your diagram and become
categories under which to list the many causes related to those
categories.
● Establish the main causes, or categories, under which other possible causes will be
listed. You should use category labels that make sense for the diagram you are
creating. Here are some commonly used categories:
● 3Ms and P - methods, materials, machinery, and people 4Ms
● 4Ps - policies, procedures, people, and plant
● Environment - a potentially significant fifth category
● Write the main categories your team has selected to the left of the effect box, some
above the spine and some below it.
● Draw a box around each category label and use a diagonal line to form a branch
connecting the box to the spine.
Step 4 - For each major branch, identify other specific factors which may be the
CAUSES of the EFFECT
● Identify as many causes or factors as possible and attach them as sub
branches of the major branches.
● Fill in detail for each cause. If a minor cause applies to more than one
major cause, list it under both.
Step 5 - Identify increasingly more detailed levels of causes and continue
organizing them under related causes or categories.You can do this by asking a
series of why questions.
Asking Questions has a great importance in drawing cause and effect diagram.
Example:
Q:Why were the TIRESUNDERINFLATED?
A: No record of tire pressure
A: Difficult air stems
Q:Why were the air stemsdifficult?
A: Poor design
NOTE: You may need to break your diagram into smaller diagrams if
one branch has too many sub-branches. Any main cause (3Ms and P, 4Ps, or a
categoryyouhave named) can be reworded into an effect.
Step 6 - Analyze the diagram. Analysis helps you identify causes that warrant
further investigation. Since Cause-and-Effect Diagrams identify only possible
causes, you may want to use a Pareto Chart to help your team determine the cause
to focus on first.
● Look at the “balance” of your diagram, checking for comparable levels of detail
for most of the categories.
• A thick cluster of items in one area may indicate a need for further study.
• A main category having only a few specific causes may indicate a need for
further identification of causes.
• If several major branches have only a few sub-branches, you may need to
combine them under a single category.
● Look for causes that appear repeatedly. These may represent root causes.
● Look for what you can measure in each cause so you can quantify the effects of
any changes you make.
● Most importantly, identify and circle the causes that you can take action on.
Detailed Exercise:
Construct a Cause-and-Effect Diagram to identify the causes of
computer downtime in your organization.

1. Draw a horizontal arrow to the right, write the effect

(computer downtime), and draw a box around it.
2. Identify the main causes contributing to the effect.
3. Draw boxes around the main categories and connect the
boxes to the horizontal arrow with diagonal arrows to form
the fishbone.
4. For each branch, identify specific factors which may be the
causes of the effect.
5. Add increasingly detailed levels of causes.
6. Analyze the diagram and circle causes that you can take
action on.
Example:
Scatter Plot Diagram
• The Scatter plot is another problem analysis tool. Scatter plots are also
called correlation charts.
• Measuring Relationships between Variables. They are often used as
follow-ups to a cause-and-effect analysis to determine whether a stated
cause truly does impact the quality characteristic.
• The scatter diagram is a plot of one variable against another variable on
a graph in an effort to see if there is a relationship between the two
variables.
Phases in preparing Scatter Diagram:
Analyze – collect and draw the data points of two variables to be related in
the x-y axis. Consider one variable as independent variable which
should be in the x-axis and the other variable as dependent variable
which should be in the y-axis.
From the data points, try to investigate the relationship between the
variables (whether it is positive or negative).

Improve (Implement) – based on the relationship between the two variables,

propose a solution for improvement.
 …..cont’d
Scatter Plot Diagram is used:
• To study and identify the possible relationship between the changes
observed in two different sets of variables.
• It immediately provides a visual means to test the strength of a potential
relationship.
• Provides a good follow-up to a Cause-and-Effect diagram to find out if
there is more than just a consensus connection between causes and
effects
• It is constructed by plotting two variables against one another on a pair
of axes.
• A Scatter plot does not predict cause-and-effect relationships. It cannot
prove that one variable causes another, but it does show how a pair of
variables is related and the strength of that relationship. Statistical tests
quantify the degree of correlation between the variables.
• For example, in an attempt to standardize inspection, several pieces
were inspected for evidence of dirt by two inspectors. The occurrences
of dirt found by each of the two inspector on the same pieces is plotted
in the figure. Note that the two inspectors are not in agreement,
inspector 2 found fewer indication of dirt than does inspector 1.
Training should be undertaken to bring the two inspectors into
agreement to give the data credibility.
 …..cont’d
Simple steps may be followed in setting up a scatter diagram:
1. Select the dependent and independent factors. The dependent factor
may be a cause on a cause and effect diagram, a specification, a
measure of quality or some other important result. The independent
factor is selected because of its potential relationship to the
dependent factor.
2. Set up an appropriate recording sheet for data.
3. Choose the values of the independent factor to be observed during
the analysis.
4. For the selected values of the independent factor, collect
observations for the dependent factor and record on the data sheet.
5. Plot the points on the scatter diagram, using the horizontal axis for
the independent factor and the vertical axis for the dependent factor.
6. Analyze the diagram.
Control Charts
• Fluctuation or variability is an inevitable component of all systems
and is expected, arising naturally from the effects of miscellaneous
chance events.
• However, variation outside a stable pattern may be an indication
that the process is not acting in a consistent manner. Events which
fall beyond expected variability or events forming a pattern that is
not random, indicate that the process is out of control.
• A control chart is therefore used to distinguish special (assignable)
causes of variation from common causes of variation, to
recognizing sources of variation.
• They are used to monitor and control a process and improve
process performance over time by studying variation and its source
on an ongoing bases.
• A typical control chart plots a selected quality characteristic, found
from subgroups of observations as a function of a sample number
(which represents the time sequence of events)
• Characteristics such as sample average, sample range and sample
proportion of nonconforming units are plotted.
• The Center line on a control chart represents the average value of
the characteristic being plotted.
• Two limits, known as the upper control limit (UCL) and lower
control limit (LCL), are also shown on control charts.
• These limits are constructed as that, if the process is operating
under a stable system of chance causes, the probability of an
observation falling outside these limits is quite small.
• Thus, the control limits are the triggers that signal when the process
is out of control.
• Samples to be taken must be random.
• Control does not necessarily mean that the product or service will
meet your needs. It only means that the process is behaving
consistently.
Sample quality
characteristic

Sample number

55
 Statistically in control vs. Technically in control
A process that is operating with only chance causes of variation present is
said to be in statistical control. Where as, a process that is operating in the
presence of assignable causes is said to be out of control.
• Statistically controlled process:
 inhibits only natural random fluctuations (common causes)
 is stable
 is predictable
 may yield products out of specification
• Technically controlled process:
 presently yields products within specification
 need not be stable nor predictable

57