SQM - Unit 3 - GRAA

SEVEN BASIC TOOLS OF QUALITY
Software Quality Management
Unit – III

Roy Antony Arnold G
Asst. Prof. / CSE
Seven Basic Tools
Seven Basic Tools
"The Old Seven." 

"The First Seven."

"The Basic Seven." 

The seven basic tools of quality are first emphasized 
by Ishikawa, a professor of engineering at Tokyo 
University.
called as father of 
He called as  father of “quality
quality circles
circles”..
SQM - Unit 3 - GRAA
Why Seven Basic Tools?
Seven Basic Tools

… The Seven Basic Tools of Quality is a designation given to a fixed 
g p q , p
set of graphical techniques, which are helpful to maintain 
software quality. 

… They are called basic because they are suitable for people with 
little formal training in statistics and because they can be used 
to solve the vast majority of quality‐related issues.

… Good use of the seven basic tools can lead to positive long‐term 
results for process improvement and quality management in 
software development. SQM - Unit 3 - GRAA
What are Seven Basic Tools
What are  Seven Basic Tools?
… The Check List or Check Sheet 
… The Pareto Diagram
… The Histogram 
… The Scatter Diagram
The Scatter Diagram 
… The Run chart (alternately Flow Chart or 
St tifi ti )
Stratification) 
… The Control Chart 
… The Cause‐and‐Effect Diagram or Ishikawa Diagram 
or Fish‐Bone Diagram
SQM - Unit 3 - GRAA
Contd...

SQM - Unit 3 - GRAA

Check List
Check List
… It is a simple document that is used for collecting data in real‐
time and at the location where the data is generated.
… I main
Its i purpose isi to provide
id a structured d way to collect
ll
quality‐related data for assessing a process or as an input to
other analyses.
analyses
… Bernstein, 1992, observed that the checklists summarize the
key points of the process and they are much more effective
than the lengthy process documents.
… Each p phase in a software development
p has a set of tasks to
complete. Checklists help developers and programmers to
ensure that all tasks are complete.

SQM - Unit 3 - GRAA

Contd...
… Another type of checklist is the common error list,
which is part of the stage kickoffs of the defect
prevention process (DPP).
… DPP involves three key steps:
† (1) analysis of defects to trace the root causes,
† ((2)) action teams to implement
p suggested
gg actions,, and
† (3) stage kickoff meetings as the major feedback mechanism

… PTF ((Program
og a Temporary
e po a y Fix)) iss tthee fix de
delivered
e ed to custo
customers
es
when they encounter defects in the software system.

SQM - Unit 3 - GRAA

Example

SQM - Unit 3 - GRAA

Example 2
Example 2

SQM - Unit 3 - GRAA

 Pareto Diagram
Pareto Diagram
… It is a frequency chart of bars in descending order; the frequency 
bars are usually associated with types of problems. 
… I i 19th Century Italian economist named Vilfredo
It is named after a 19
d f C I li i d Vilf d
Pareto (1848–1923), who expounded his principle in terms of the 
distribution of wealth—that
distribution of wealth that a large share of the wealth is owned 
a large share of the wealth is owned
by a small percentage of the population. 
… In 1950 Juran applied this principle to the identification of quality 
In 1950 Juran applied this principle to the identification of quality
problems—that most of the quality problems are due to a small 
percentage of the possible causes. 
… Pareto analysis is commonly referred to as the 80–20 principle (20% 
of the causes account for 80% of the defects)

SQM - Unit 3 - GRAA

Contd...
… In software development, the X‐axis is usually the defect 
cause and the Y‐axis the defect count. 
… I i di
It indicates which problems should be solved first in 
hi h bl h ld b l d fi i
eliminating defects and improving the operation. 
… G d
Grady and Caswell 
dC ll (1986) shown a Pareto analysis of 
(1986) h P l i f
software defects by category for four Hewlett‐Packard 
software projects
software projects
… The top three types (new function or different processing 
required existing data need to be organized/ presented
required, existing data need to be organized/ presented 
differently, and user needs additional data fields) account for 
more than one‐third of the defects.
SQM - Unit 3 - GRAA
Pareto Analysis for IBM Rochester 
product
d

INTF Interface Problems

INIT Data Initialization Problems
CPLX Complex Logical Problems
NLS Translation related
National Language
g g Problems
ADDR Address Problems
DEFN Definition Problems

SQM - Unit 3 - GRAA

SQM - Unit 3 - GRAA
Histogram
… It is a graphic representation of frequency counts of a 
It i hi t ti ff t f
sample or a population. 
… Th X i li t th
The X‐axis lists the unit intervals of a parameter (e.g., 
it i t l f t (
severity level of software defects) ranked in ascending 
order from left to right and the Y axis contains the
order from left to right, and the Y‐axis contains the 
frequency counts. 
… The purpose of the histogram is to show the distribution
The purpose of the histogram is to show the distribution 
characteristics of a parameter such as overall shape, 
central tendency dispersion and skewness
central tendency, dispersion, and skewness. 
… It enhances understanding of the parameter of interest.

SQM - Unit 3 - GRAA

SQM - Unit 3 - GRAA
Customer Satisfaction Histogram
Customer Satisfaction Histogram

SQM - Unit 3 - GRAA

Scatter Diagram
Scatter Diagram
… A scatter diagram vividly portrays the relationship of two interval 
variables. 
… Compared to other tools, the scatter diagram is more difficult to 
Compared to other tools the scatter diagram is more difficult to
apply. 
… It usually relates to investigative work and requires precise data. It 
t usua y e ates to est gat e o a d equ es p ec se data t
is often used with other techniques such as correlation analysis, 
regression, and statistical modelling.
… Each point in a scatter diagram represents an observation of both 
the dependent and independent variables. 
… SScatter diagrams aid data‐based decision making (e.g., if action is 
tt di id d t b d d i i ki ( if ti i
planned on the X variable and some effect is expected on the Y 
variable). 
SQM - Unit 3 - GRAA
Contd...
… One should always look for a scatter diagram when the 
correlation coefficient of two variables is presented.
… Th
The method for calculating the correlation coefficient is 
h df l l i h l i ffi i i
highly sensitive to outliers, and a scatter diagram can clearly 
expose any outliers in the relationship
expose any outliers in the relationship. 
… The most common correlation coefficient is Pearson's 
product moment correlation coefficient, which assumes a
product moment correlation coefficient, which assumes a 
linear relationship. 
… If the relationship is nonlinear, the Pearson correlation 
p ,
coefficient may show no relationship; therefore, it may 
convey incorrect or false information.

SQM - Unit 3 - GRAA

Correlation of Defect Rates of Reused Components 
B t
Between Two Platforms
T Pl tf

SQM - Unit 3 - GRAA

Run Chart
Run Chart
… A run chart tracks the performance of the 
parameter of interest over time. 
… These charts serve as real‐time statements of 
quality as well as workload.
… The X‐axis is time and the Y‐axis is the value of the 
parameter. 
parameter.
… A run chart is best used for trend analysis, especially 
if historical data are available for comparisons with
if historical data are available for comparisons with 
the current trend. 

SQM - Unit 3 - GRAA

Contd...
… Ishikawa (1989) includes various graphs such as the
pie chart, bar graph, compound bar graph, and
circle graph under the section that discusses run
charts.
… An example of a run chart in software is the weekly
number of open problems in the backlog; it shows
the development team's workload of software
fixes.

SQM - Unit 3 - GRAA

Run Chart of Percentage of 
Delinquent Fixes
l

SQM - Unit 3 - GRAA

Control Chart
Control Chart
… The control chart is a powerful tool for achieving 
Th l h i f l lf hi i
statistical process control (SPC). 
… H
However, in software development it is difficult to 
i f d l i i diffi l
use control chart in the formal SPC manner. 
… I i f
It is a formidable task, if not impossible, to define 
id bl k if i ibl d fi
the process capability of a software development 
process. 
process
… A control chart can be regarded as an advanced 
form of a run chart for situations where the process
form of a run chart for situations where the process 
capability can be defined. 

SQM - Unit 3 - GRAA

Contd...
… It consists of a central line, a pair of control limits (and 
sometimes a pair of warning limits within the control limits), 
and values of the parameter of interest plotted on the chart
and values of the parameter of interest plotted on the chart, 
which represent the state of a process. 
… The X‐axis
The X axis is real time. If all values of the parameter are 
is real time If all values of the parameter are
within the control limits and show no particular tendency, the 
process is regarded as being in a controlled state. 
… If they fall outside the control limits or indicate a trend, the 
process is considered out of control. Such cases call for 
causal analysis and corrective actions are to be taken.

SQM - Unit 3 - GRAA

Contd...
… In statistical terms, process capability is defined:
USL − LSL
CP =
6 Sigma
… where USL and LSL are the upper and lower engineering
specification limits, respectively, sigma is the standard
deviation of the process, and 6 sigma represents the overall
process variation.
i i
… If a unilateral specification is affixed to some characteristics,
th capability
the bilit index
i d may be b defined:
d fi d (
(u‐process mean))
USL − u u − LSL
CP = CP =
3Sigma
SQM - Unit 3 - GRAA
3Sigma
Pseudo‐Control Chart of Test Defect Rate—First 
It ti
Iteration

SQM - Unit 3 - GRAA

Example: 2
Example: 2

SQM - Unit 3 - GRAA

Cause‐and‐Effect
Cause and Effect Diagram
Diagram
… This was developed by Ishikawa and associates in the early 
1950s in Japan. 
… I
It was first used to explain factors that affect the production 
fi d l i f h ff h d i
of steel.
… It sho s the relationship bet een a quality characteristic
It shows the relationship between a quality characteristic 
and factors that affect that characteristic. 
… Its layout resembles a fishbone
Its layout resembles a fishbone.
… While the scatter diagram describes a specific bivariate 
relationship in detail the cause‐and‐effect diagram identifies
relationship in detail, the cause‐and‐effect diagram identifies 
all causal factors of a quality characteristic in one chart.

SQM - Unit 3 - GRAA

Cause‐and‐Effect
Cause and Effect Diagram
Diagram
… These are diagrams show the causes of a certain event 
Th di h th f t i t
… Common uses of this diagram are product design and quality 
defect prevention, to identify potential factors causing an 
p , yp g
overall effect. 
… Causes are usually grouped into major categories to identify 
these sources of variation The categories typically include:
these sources of variation. The categories typically include:
† People
† Methods
† Machines
† Materials
† Measurements
† Environment

SQM - Unit 3 - GRAA

Cause‐and‐Effect
Cause and Effect Diagram
Diagram

SQM - Unit 3 - GRAA

Cause‐and‐Effect Diagram Contd.

… Typical categories of causes are:

Typical categories of causes are:
† The 8 Ms (used in manufacturing)
„ Machine (technology) 
„ Method (process)
Method (process) 
„ Material (Includes Raw Material, Consumables and Information.) 
„ Man Power (physical work)/Mind Power (brain work)
„ Measurement (Inspection) 
„ Milieu/Mother Nature (Environment) 
„ Management/Money Power 
„ Maintenance 
† The 8 Ps (used in service industry)
The 8 Ps (used in service industry)
„ Product=Service  ▪ Price 
„ Place  ▪ Promotion/Entertainment 
„ People (key person)
People (key person)  ▪ Process 
Process
„ Physical Evidence  ▪ Productivity & Quality 
† The 4 Ss (used in service industry)
„ Surroundings 
g
„ Suppliers 
„ Systems  SQM - Unit 3 - GRAA
„ Skills