Sunteți pe pagina 1din 46

Module 2

Quality Management

Topics in this module

Concept, Dimensions and Definition of quality
Cost of quality

Quality Assurance and ISO 9000

Total Quality Management TQM
Statistical Quality Control
Control Charts
Six Sigma Approach

Concept of Quality

The word Quality has different meaning, concept and perception to different people.
Some of them are
- Fitness to use
- Compliance with specified requirement
- Free from defects, imperfection and contamination
- Degree of excellence
- Level of Customer satisfaction by way of value for the money paid etc

Dimensions of Quality

For a Manager, Quality is judged or assessed by a set of Dimensions of Quality and

not by perception

The Dimensions of Quality are different for a Product and a Service.

1 Dimensions of Product Quality

1. Performance Main characteristics of the product which it is expected to

perform. Eg. Washing Powder is expected to clean cloths quickly

2. Special features special features of the product which differentiate it from

competing products. Eg. Fragrance to the washed cloth, protect colours

3. Reliability consistency of performance year after year eg. packets purchased

every month have same quality
4. Conformance product compliance with design specifications. Eg. Accuracy of
weight, composition etc

5. Durability useful life of the product eg. Washing machine to last trouble free
for min 5 years or till technology changes eg. CRT to LCD/LED TVs

6. Aesthetics appearance, feel, smell, taste of products and packing eg. Colour
and shape of refrigerator

7. Perceived quality Quality associated with reputation and brand name eg.
TATA products or V Guard are expected have good product quality

8. Serviceability Adequate servicing network to handle customer complaints,

repair etc. Eg. Large no. of Maruti service centres to handle customer
complaints compared to Ford motors

2 Dimensions of Service Quality

1. Tangibles
Physical appearance of sales and service office, its facilities and its employees, their
communication capability. Eg. Airline ticketing office

2. Convenience
Accessibility and availability of services nearer to the customer eg. Nokia mobile
phone service centres

3. Assurance
Competence and ability to convey trust and confidence of service personnel eg. A
Samsung service technician is capable to service all Samsung household products

4. Reliability
Dependency of service providers for delivery of service eg Courier Service co,
logistic companies, Authorized service centres of LG

5. Time
Speed at which service is provided- quick response time to a service request. 24 Hr
on the road mobile servicing units of Hyundai Motors

6. Responsiveness
Willingness to help customers in unusual circumstances to solve their problems eg.
insurance companies settling complex claims quickly, air lines providing ticket for
emergency travel
7. Courtesy
The way customers are treated by employees of service providers eg. cabin crew of
air line, single window service in banks

ISO Definition of Quality

Quality is defined as the Totality of characteristics of an entity that bear on its
ability to satisfy the stated and implied needs.

means features of product or services as stipulated in product catalogue or service

- refer to a particular product or service

Stated need
- needed to be satisfied as per catalogue or contract

Implied needs
- needs expected to be satisfied e.g. Washing soap

Cost of Quality

1. Internal failure cost

It is the cost associated with in the company in the flowing areas
o Cost of rejected parts during manufacturing and assembly material cost,
labour cost and overhead
o Cost of rework labour cost for correction of defective parts
o Scrap and rework of purchased parts cost to the buyer to resolve the quality
problems of purchased part
o Cost of doing 100 % inspection due to poor quality of parts beyond acceptance
level when doing sampling inspection
o Re-inspection cost cost of re- inspecting reworked and re-supplied items
o Lost and missing information cost of search and retrieval eg. Inspection report
of a sub assembly
o Cost of unsold products due to change in quality requirements of customers
o Cost of down grading selling poor quality good at lower price

2. External Failure cost

Cost associated with defect found at customers place.
o Warranty Cost Cost of repair and replacement Mat + Lab + OH
o Complaint handling cost cost of investigation and adjustments done to
overcome the defect
o Return Management cost cost of taking back defective parts and products
o Allowances cost of concessions given to customer due to poor quality eg.
extended warranty, price discount etc.
o Penalty cost penalty for defective products or services
o Revenue loss Delayed or reduced payment due to defects in products sold
o Lost opportunity revenue loss of existing customers and new customers

3. Appraisal cost
Cost to determine the degree of compliance to quality
o Incoming inspection cost for purchased materials and parts
o In process inspection cost for manufactured and assembled parts
o Documentation cost paper work and record keeping of inspection reports
o Equipment cost of inspection and testing equipments, lab etc
o Evaluation cost evaluation of stocked materials and products for determine

4. Prevention Cost
Cost of activities to prevent failure and to reduce appraisal cost
o Quality planning cost of preparing quality procedures and systems
o New product review cost of reliability engineering and product review
o Process capability planning and review to ensure that the manufacturing
processes and systems gives the required quality.
o Quality Audit cost of auditing existing systems
o Supplier quality evaluation evaluate and improve supplier quality during
contract period.
o Training cost for training of employees on quality

5. Hidden Cost
Cost that cannot be quantified or measured accurately
o Loss of revenue and profit from existing and new customers
o Lower market share market being taken over by competitors
o Loss of corporate image affecting credit and investment flow
o Unreported failure cost due to fear of punishment
o Poor employee morale having to a work again and again
o Life cycle cost higher cost of stocking & supplying spares and services in the
entire product life

Quality Assurance

Quality Assurance is fundamentally different from Quality Control/Inspection

Quality Assurance is a pre/during production activity and
Quality C is a post production activity
Control loop of Process and Quality Assurance is as given below

Input Process Output

Standards &
Process Quality Design
control Monitoring documents


Materials /labor

o Any process have Input and Output

o The output quality is monitored continuously or intermittently with reference to
set standards and design drawings
o If there is a deviation against standard, the process is set right through process
control (eg resetting the RPM of drilling machine)
o If the deviation is continuing, the problem is solved by machine maintenance or
changes in material or labour input till the process is back to acceptable level.
o This system is continuous closed loop system and will assure that what ever is
produced is of acceptable quality. This is called Quality Assurance, which will
ensure quality in every process in the factory.
Difference between Quality Assurance and Quality Control (Inspection)
Criteria Quality assurance Quality Control/Inspection
Focus Customer focus - meeting Product focus meeting
quality needs of the quality needs of the product as
customers per design

Process Pre-production process Post production process

quality is built into the quality is controlled by
production process itself inspection of the items after

Goal Long term customer Short term profit generation


Culture Quality is a culture of the Quality is only a functional

organization and business requirement

Cost Eliminate rejection and High cost of rejection and

rework rework

Participation Full participation of all Participation limited to their

employee in every quality own work

Quality Assurance and Selection of Process (Process Choice)

- Quality of a product is closely associated with the Process Capability of the

machinery & equipment

- Process Capability is the capability of the machine to produce parts with a

necessary quality.

- Low cost machine will have low quality output

- high cost will have high quality output
- Quality of the part produced is monitored against standards and design drawings
- Cost of Quality assurance consists of
- cost of continuously or intermittently monitoring the process
- cost of rejection and rework
- cost of frequent resetting or repair of machine to get the desired quality
- cost of maintaining necessary records of parts inspected
- Cost of Process choice Vs Cost of Quality Assurance is given below.

Cost Total cost

Optimum choice of process (m/cs)
Cost of
(cost of m/cs)

Cost of machines
Low processHigh process Cost of Q.A.
Capability capability
(High rejection)(Low rejection)

- The two costs associated with Operations are Cost of QA and Cost of Machines
- With increasing cost of a machine (ie Cost of Process Choice),the cost of QA
decreases and cost of production increases.
- The sum of these costs is the Total cost which takes a U shape.
- Hence the machine to be purchased (Investment in Process Choice) should be of
such Process Capability (quality of output) such that the total cost is minimum.
Eg. An ordinary drilling machine that cost Rs 1 lakh only is needed for drilling open
dimension holes where as high quality drilling machine costing Rs 4 lakh is needed for
drilling precision holes.

Vice versa will only increase total cost of production

ISO 9000 Quality Standards


ISO stands for International Organization for Standardization. This organization was
formed in Geneva, Switzerland in 1946.

Objective of ISO 9000

1. Objective is the establishment of a Quality Management System (QMS),
which has international acceptance for the benefit of both the suppliers and the

2. Quality Management System essentially means that quality is built into the
system of various Management processes, Production processes and
Procurement processes so as to ensure the required quality to the end product is
maintained and improved on a continuous basis.

3. ISO 9000 QMS is different from inspection or quality control because

inspection is a post production activity.

4. Quality control is only one part of ISO 9000.

5. In ISO 9000, the processes and systems will assure quality, because quality is
built into the process itself

6. ISO 9000 is a tool to achieve, sustain and improve quality

7. Establishment of ISO 9000 is essential for to achieve higher levels of

operational performances using TQM. TPM, ERP etc

Systems and Elements of ISO 9000

ISO 9001 comprise of 3 Systems and 20 Elements

1. System Management

2. System Methodology

3. System Maintenance

1 System Management (4 elements)

Elements are

1. Management responsibility clearly defined quality responsibility at all levels

2. Quality system documented inspection procedure, calibration of measuring


3. Corrective and preventive action documented actions

4. Internal quality audit verification of ISO compliance by the company itself

2 System Methodology (13 elements)

Elements are

5. Contract review compliance as per order placed by all customers w.r.t. time,
delivery, quality, performance

6. Design control to ensure that products are produced as per the latest design which
are modified from time to time to meet changing customer needs.

7. Document and data control maintenance of all data and records for traceability of
the reason for any defect which can appear at a later date while in use.

8. Purchasing system to ensure the required quality at the vendors production

processes to assure

9. Control of customer supplied product if such situation occur

10. Product identification and traceability documentation for tracing the source and
reasons for any defect found at a later date in the product

11. Process control ensure production machines are capable of giving the required

12. Inspection and testing clearly defined inspection and testing methods and

13. Control of non-confirming product because of very high cost, certain parts are
not rejected due to deviation in certain dimensions, but other parts are modified to give
the desired performance this is a non-confirming product documentation and
performance monitoring and control of such products

14. Handling, storage, packaging and delivery a documented system for all these

15. Training documented requirements of employee training and training imparted

16. Servicing documented servicing procedures and servicing records

17. Statistical techniques different statistical tools used for defect/deviation analysis

3 System Maintenance (of QMS)( 3 elements)

Elements are
18. Control of testing, measuring and inspection equipments documentations for all

19. Inspection and test status records of all inspection

20. Control of Quality Records preservation of all quality records for future
reference with clearly assigned responsibilities

ISO 9000 Series of Quality Standards

ISO 9000 series have 5 Main Documents. They are ISO 9000, 9001,9002,90003, 9004

1. ISO 9000

These standards provide guidelines for selection of ISO 9000 family of standards (ISO
9001, 9002 or 9003) as applicable to the particular business.

2. ISO 9001

This Standard provides Quality standards as applicable to the companies and

businesses engaged in Design, Development, Production, Installation (including
testing and commissioning) and Servicing. Almost all engineering industries come
under ISO 9001 series. Eg. BHEL, Tata Motors, Maruti, HMT

3 ISO 9002

This document gives Quality standards as applicable to the companies and business
engaged only in production, installation/supply and servicing. (No product design
and development is undertaken by the company)

Most of the process based industries in petroleum, fertilizer etc where the company by
itself do not carry out any product design and product development work, come under
this series.

E.g. Urea produced by FACT and other companies are same in quality and
composition. They do not carryout designing of Urea. ISO 9002 is applicable to
FACT Urea Plant

4 ISO 9003

These standards relate to final inspection and testing system and procedures for both
in-house production as well as purchased parts and materials.

5 ISO 9004

These standards relate to internal Quality Assurance and Quality Audit.

It deals with quality in documentation and corrective and preventive actions.

Indian Standard Equivalents of ISO 9000 series

ISO 9000 to ISO 9004 series of standards are represented by and equivalent to IS
14000, 14001, 14002, 14003 and IS 14004 series of Indian Standards respectively.

15 steps of ISO 9000 implementation

1. Top management commitment

- formulate quality policy

- set quality objectives of implementation

- provide leadership and funds for ISO implementation

2. Appoint Consultant

- Consultant will help preparation of Quality Manual and other ISO 9000

- He will provide overall guidance for implementation

3. Nominate Management Representative (MR)

- MR will hold overall responsibility of implementation and maintenance.

- Top management should identify an executive as MR

4. Constitute Implementation team

- Implementation team consists of members from different functional areas

- Implementation team will assist ISO implementation in different functional

5. Training and awareness

- impart company wide training and awareness program to each and every
employee on various aspects and requirements of ISO 9000. It normally take
about an year for a medium sized company

6. Finalize Time Schedule

- Prepare and publish a time schedule for different implementation activities

across the company

7. Element Owners

- Identify executives and fix responsibility for each of the 20 elements of ISO

- These Element Owners will ensure ISO implementation and maintenance as per
the Company Quality Manual.

8. Review present system

- Each department will review present system with the assistance of

Implementation Team.

9. Prepare Documentation

The following ISO 9000 Quality Documents are prepared by MR, Implementation
Team, Consultant, Element Owners and Departmental Executives

- Company Quality Manual

- Departmental Manual

- Working Instructions

- Various reporting Formats, Forms etc with Identification No. eg. Purchase
Order format, Production Completion Report, Inspection report etc

10. Install new system

- Install new system in all departments with effect from certain date

11. Internal Audit

- Conduct internal audit by the Internal Audit Team after 6 months of

implementation as per Internal Audit Procedure specified in Quality Manual
- Identify areas of Non- Conformance with the ISO System and prepare Non-
Conformance Report (NCR)

12. Management Review

- Review all NCRs and the effectiveness of implementation with reference to

stated Objectives in the Quality Manual

- Prepare corrective and preventive action plan and implement them

13. Pre- Assessment audit

- Pre- assessment audit is done by a team of External Auditors, accredited to the

ISO 9000 Certifying Agency.

- Internal Auditors will assist External Auditors in Pre-Assessment Audit.

- If Pre assessment audit is successful (absence of any major NCRs), the

company is permitted to apply for ISO 9000 Registration and Certification
Audit with the Certifying Agency

14. Registration and Certification Audit

- Company has to register with the ISO 9000 Certifying Agency for certification
by paying necessary fees.

- They will send team of External Auditors to do the Certification Audit

- If there is no major NCR, Certification Auditors will recommend issue of ISO

9000 certificate to the company.

15. Award of Certification

- Based on these recommendations, ISO 9000 Certificate is issued by the

Certifying Agency.

Maintenance of ISO 9000 Certification

- Internal Audit is conducted by the company usually every 6 months, but before
Surveillance Audit by External Auditor

- Non Conformance Reports from Internal Audit is reviewed by the Management


- Corrective and preventive actions are decided, implemented are monitored.

- Every six months, Surveillance Audit is done by the certifying body auditor.
- If the system is working OK, and there are no major Non- Conformities,
company continues to remain as ISO 9000 certified.

Advantages of ISO 9000

Following are the major advantages of implementing ISO 9000 system

1. Quality is built into the system, i.e. the quality is assured if the systems are
correctly followed.

2. The dependence on traditional inspection and quality control is very less.

3. Customers become more confident of the products and services of the company.

4. Provide competitive edge in domestic and international market.

5. Provide climate and scope for continuous quality improvement.

6. Reduce waste and increases productivity.

7. Increases profit.

8. Bring about attitudinal change in the employees, resulting in their participation

and better performance.

9. Each employee feels his importance and role in the operations of the company.

10.Encourage innovation and improvements in all areas.

11.ISO 9000 provide the foundation for implementing advanced performance

improvement systems like TQM, TPM, MRP, JIT, BPR, ERP, SCM, CRM etc.

Problem faced during ISO 9000 Implementation

1. Lack of management Commitment

- ISO 9000 is a top down management activity. Top management and all
executives should be fully committed to the system to make it a success. Any
laxity on their part will make the system ineffective.

2. Lack of resources
- Implementation is very challenging and demanding to all resources, including
money, manpower and time. Companies with out these resources will face

3. Resistance to change

- Change of work culture and systems will face stiff resistance from several
employees and managers in the beginning.

- This is mainly because of the human nature to resist changes. This needs careful
handling by bringing in high awareness in each and every employee at all

4. Capital Investment

- Capital Investments are essential for upgrading inspection and testing facilities
and some production facilities to meet the quality requirements. With out capital
investment, ISO 9000 cannot be maintained in long run

5. Lack of focus on ISO 9000

- Continuous maintenance and improvement of the system after Certification

need constant attention and monitoring from management.

- With lack of focus, the company will slip back to old work habits resulting in
wastage of the money and efforts for certification.

Total Quality Management (TQM)

ISO definition of TQM

TQM is defined as a management approach of an organization centered on quality,

based on the participation of all its members, and aiming at long term success
through customer satisfaction, and benefit to all members and society.

Components of TQM
Leadership and
Management Total
Commitment Customer

Human Resources,
Resources & Systems and
Team work Techniques

ISO 9000 Foundation

TQM Model
Fig. 15.4- 1

1. It has a Quality Management System (ISO 9000) as its foundation.

2. It is a top down management approach with full commitment and leadership

from top management.

3. It is supported by physical Resources (4M), Systems and Technologies for

continuous improvement at all levels. eg. good machines, tools, materials,
quality assurance systems, new technologies etc.
4. Human resources and team work are considered vital and having unlimited
potential with their participation to increase customer satisfaction.

5. The corporate focus is to maximize Total Customer Satisfaction.

Elements of TQM or 8 Pillars of TQM

1. Continuous Improvement

2. Competitive Bench Marking

3. Employee empowerment

4. Team Approach

5. Fact based decisions

6. Knowledge of quality tools

7. Suppliers/Vendors Quality

8. Quality at the Source

1. Continuous Improvement

TQM aim at continuous improvement of materials, methods, machines and

people even if they are presently working well. The term Kaizen refer to
continuous improvements even in small steps, a technique perfected by Japan to
achieve global competitiveness.

2. Competitive Bench Marking

Identify other companies to benchmark, study their way of doing and improve
your operations. Xerox benchmarked mail-order company L.L. Bean to improve
Order filling.

3. Employee empowerment

Motivate and encourage innovation. Give workers the authority to make

changes in their operational area and make them responsible for the
improvement of performance.

Eg. reduction of production time by 5% with their own innovative ideas.

4. Team Approach

Use synergy from team work for problem solving

5. Fact based decisions

Decisions based on data and its analysis rather than opinions.

6. Knowledge of quality tools

Knowledge of Tools of TQM by all employees and managers for data analysis
and decisions

They analyse quality problems using TQM tools and arrive at a solution by
7. Suppliers/Vendors Quality

Treat Suppliers of raw materials and parts as long term partners with high
concern for quality and customer satisfaction. Extend TQM to all suppliers to
enhance quality

8. Quality at the Source

Follow principle of Do it right first time and every time. Make the doer
(worker) responsible to the next internal customer with authority to use
resources provided to him. Do away with Inspector regime as far as possible

Traditional Vs TQM approaches

Aspect Quality approach TQM
Over all Maximize Return on Meet or exceed customer
mission Investment satisfaction

Objective Emphasis on short term Balance short term and long


Focus Product oriented Process oriented

Management Inconsistent objectives, not Consistent objectives, Open,

always open encourage employee

Role of Issue orders, enforce the Coach, build trust, remove

manager order barriers

Customer Unclear, not always top High priority, important to

requirements priority identify customer needs

Problem Blame, unsystematic, Identify, resolve by team

solving individualistic work

Improvemen Erratic Systematic and continuous

Adversaries Partners
Narrow, specialized, Broad, general, much team
Jobs individual effort work
Advantages of TQM
1. Companies achieve high competitiveness in cost, quality and
customer satisfaction.

2. Result in high employee motivation, innovation and

participation to be competitive.

3. Facilitate continuous product, quality and cost improvement as

per customer needs.

4. Attain capability to expand in national and international market.

5. Develop excellent quality culture in the organization, reflecting

quality in all spheres of activities.

6. Develop internal strength through its employees and systems to

overcome bad times and competition.

Problems faced in TQM Implementation (reasons for failure)

1. Lack of awareness lack of company wide awareness of TQM
requirements and culture of working leads to ineffective

2. Resistance to change employees resist changes and avoid

taking responsibility for the quality in their work.

3. Additional responsibility with out adequate authority make

people responsible with out adequate authority to change and
manage their resources to achieve the required performance

4. Changing management priority -Priority for quantitative out put

instead of qualitative output to meet production targets.

5. Lack of monitoring loss of importance to TQM from top

management due to difficulty business environment leads to its

6. Inadequate knowledge of Tools of TQM by all for problem

7. Inadequate remuneration - Additional responsibility for problem
identification and solving without additional remuneration to

Seven Tools for TQM

A number of tools are available to aid management to interpret and

analyze data and take decisions. Some of them are

1. Flow Charts


Flaws Defective

- Flow charts are visual representation of processes.

- Diamond shapes represents decision points on flow chart.

- Rectangular shapes represent activities or procedures.

- It will help to identify where exactly the problems occur.

- Very useful to analyze long processes with several activities/

procedures and decision points to make the process more

- Eg. Salary payment process, Order fulfilment process

2. Check Sheets
It gives a format to record certain data in a structured manner.
It is a simple tool used for problem identification and analysis
Rejection data of 4 parts with Batch Qty of 1000 each
Date Part A Part B Part C Part D
Rejectio rejectio rejectio Rejs
ns ns ns
01.01.1 3 5 2 -
0 1 2 2 2
02.01.1 4 2 1 -
0 1 5 - -

The above data immediately draw your attention to Part B which has
more rejection and need corrective actions

3. Histogram
Histogram is a graphical representation of distribution. It will provide
clear visual idea of distribution and range.
It will also highlight unusual values in a given data.
Eg. Histogram of Machine Breakdown causes A to F and their
frequency of occurrence.


Machine Break down causes

The histogram shows, C is a major cause of break down and F is a

minor cause
4. Pareto Analysis
Pareto analysis is the technique of focusing on the most important
problem area and resolve that problem on priority.


Machine Break down causes

Popular 80-20 rule is based on Pareto concept, which says 80 % of

problems is contributed by 20% of the items and vice-versa.
Eg. 80 % of product defects come from 20% of causes.
A Pareto diagram is used for Pareto analysis.
Eg. Pareto diagram for machine breakdown . Major causes for
machine break down are 2 (C&D) out of total 5 causes. They are to
be tackled immediately to reduce break down time substantially.
5. Scatter Diagram
Scatter diagram show the degree and direction of relationship
between two variables.

E.g. The relation ship between % of rejection while machining a

component and the humidity of the room. Scatter diagram shows
rejections are higher when humidity is higher. Action is to reduce
humidity to reduce rejection.
* *
% of rejection *


* *
% Humidity

The rejections can be reduced by maintaining humidity at lower

6. Control Charts
Different process control charts (X bar, R, p , c charts etc) are used
to monitor the Production Process out put to ensure that it is
between the Upper and Lower Control Limits with respect to certain
parameter e.g. quality

* * * * *
Quality * * * *
* * * * *
Parameter * *
Eg. Hole Dia LCL
* *

Lot No

Lots where the value lies between UCL & LCL shows the process is
under control. Variations are random or natural deviations .
When the value goes beyond UCL and LCL, the process is not under
control and the variations are due to certain causes
7. Cause and Effect Diagram or Fish bone diagram
It is a method to search for possible causes for a problem or process
It is a systematic problem solving tool.
Possible Sources and Causes of problems are explored and examined
through group discussions by all concerned.
Detailed analysis is made by asking who, what, where, when, why
and how to find the source and root cause of the problem.
Eg. Poor Surface finish (Effect) in a Grinding operation

Supervision Environment Materials (Source)

Composition (Cause)
No Supervisor
Humidity Hardness
Soft wheel Surface finish
Not trained
Vibration (Cause)

Statistical Quality Control (SQC)

- SQC is an application of a set of statistical tools for controlling
and improving quality where ever required in the whole

- It is not only applicable in production but also in other

functional areas like finance, marketing, sales and service, HR,
purchase etc.

- In high volume production, it is not possible to inspect all the

parts/materials produced or purchased.

- i.e. 100 % inspection is not possible, it is time consuming,

expensive and non-productive

- Hence Statistical Quality Control (SQC) is used to ensure that

the products and parts have the required quality.

There are 3 categories of SQC techniques

1. Traditional Descriptive statistics
2. Statistical Process Control
3. Acceptance Sampling
1. Descriptive statistics Normal Distribution Curve
- This is a post production process of quality measurement

- The frequency distribution of the measurements of all parts

produced follow a normal distribution curve as shown below

Normal Distribution

Mean + 3


The measurements will have a Mean Value, Standard deviation and

Range as given blow

Mean x x / n

(x x ) 2
Standard deviation =

Where x = each measurement

n = number of measurements
Range = Difference between the maximum vale and minimum value
of the lot
For a data that come under normal distribution, it can be said that
the Probability of Occurrence associated with the measurement
is as follows.
99.75 % of values normally fall with in the x 3
95.45 % of values normally fall with in the x 2
68.26 % of values normally fall with in the x 1
These % are called Confidence Level of the data
( ie, Nearly 5 defective parts in 100 parts produced under x 2 . Or
only 95 % confidence level that the parts are good)
In Six Sigma approach x 6 means Very high Confidence level to
assure that only 3.4 defects are there in I million parts

2. Statistical Process Control

- this is done during the production process to ensure that
the production process is statically under control to ensure
production of only good parts.

Input Process Output

Standards &
Process Quality Design
control Monitoring drawings


- Quality of a product has great dependence on the quality conformance of the


- It means the out put of the process confirm to the intended design of the
process output (eg. a grinding machine gives x micron surface finish and y
micron concentricity).

- Since it is not possible to inspect 100 % items, the quality is assured statistically
from samples taken at intervals during the process

- Statistical Process Control is a tool to evaluate the output of the process and
ensure that it is statistically acceptable.

- Different Control Charts ( x bar, R, c, p chart etc) are used for process control

3. Acceptance Sampling
- It is used to inspect and accept parts and raw materials purchased from suppliers
- It is also used for inspection and acceptance of finished product from assembly
line (a post production process)
- OC (Operating Characteristics) Curve is used to accept Purchased Parts
- AOQ (Average Outgoing Quality) graphs are used for quality analysis and
control of finished products produced eg. Electric Fans going out of a company

Statistical Process Control

In any process, 2 parts cannot be made with exact identical dimensions.
- eg. dia 20 mm hole made in the first piece and the second piece will be different
in diameter, one may be 20.01 mm and the other may be 19.98 mm. and not
exactly identical.
- This difference is called Process Variations

Types of Variations
1. Random variations or Chance Variations
o Variations occur at random
o They are caused by complex minor causes
o do not result in significant deviation from the nominal value eg dia 20.01,
19.99, 20.02 etc
2. Assignable variations
- variations are relatively large e.g. dimensional deviations of a machined part
(dia 20.2, 20.8. 19.5 etc), chemical composition deviation of chemical product.
- Causes of variations can be assigned to one or more of the following causes
o Due to workmen
o Due to machine
o Due to materials
o Due to tools
o Due to environment like humidity, dust etc
o Or combination of any of the above

Process Variations are represented and controlled using different Control Charts
Control Charts
Types of control charts
1. Control chart for Variables: variables are based on measurements like
dimensions, temperature, rpm, surface finish, chemical composition etc (x
bar and R Chart)

2. Control charts for Attributes: attributes are based on Numbers. The

population is classified into two classes, right or wrong, good or bad,
without errors or with errors

e.g 92 nos of invoices without error and 8 with error in 100 invoices checked
(p chart)
e.g. 32 number Mobile Phones with total 5 or more defects vs 9968 number
with less then 5 total defects. (c chart) in a population of 10000

Control Chart for Variables

Control chart for Variables with Periodic Inspection data (Construction of

Control Charts)

If the cycle time for a process is more (eg. 5 to 10 mts), then one sample is taken for
inspection in every 30 to 60 mts in a shift. Control Charts are prepared based on this
inspection data
1. x Chart (x bar chart)

Periodic Measurements are x1, x2, x3, .., xn

N = number of measurements or samples taken
Mean x ( x1+ x2 + x3 +..+ xn) / n

(x x )

Standard deviation =
Upper Control Limit UCL = x + 3
Lower Control Limit LCL = x - 3

* (Assignable cause)
Quali ty
Parameter UCL
* *
* * *


* *
* * LCL
* (Assignable Cause)

123 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Number

Interpretation of Control Chart

The variations in dimensions of samples which fall with in the UCL
and LCL are due to random variations and are acceptable
The variations in dimensions of samples which fall outside the UCL
and LCL due to assignable variations reasons for which are to be
investigated and corrected to bring the process under control ie.
within acceptable limits .

Construction of Control Charts

(Establishing revised Mean, UCL and LCL for the Process)
If some values are falling outside the calculated UCL and LCL, these
values are deleted from the data.
Then a revised mean, UCL and LCL are calculated.

A new Process Control Chart is drawn and this represents the


As the process continues for the next shift and subsequent days,
samples are taken at certain intervals.
If the values are within New UCL & LCL, we can cay that the process
is under control with 99.75 % confidence level (corresponding to +-3
sigma level)

Control chart for Variables from Sampling Inspection data

When the production rate and production volume is high (like a Soft Drink Bottling
Plant eg. One bottle in every 10 seconds), instead of Single Piece sample, 2 to 5 pieces
(sample size) are taken for inspection in each lot at an interval of 30 minutes (lot no.)

Then control charts are prepared based on sampling Inspection

2. x Chart (Mean Chart)
It is assumed that frequency distribution of sample lot is similar to the frequency
distribution of the population.
The sample size (n) and number of lots (K) taken per shift for inspection, is based
on a Sampling Plan, accuracy requirements etc
n = sample size
K = Number of lots taken for analysis
x = Sample Mean, ie average of each sample = (x1 + x2 + xn)/n
x1, x2, xn = values of each piece in one sample on n pieces
x = mean of the Sample Means
= x / K

R = Average of sample ranges

= R /K

R = sample range = Difference between the maximum and minimum values is a


UCL and LCL of x chart = x A2 R

A2 is constant for the sample size n at a certain confidence level ( value of
constants are available in A2 table for different sample sizes and confidence levels
refer any OM Book)
* (Assignable cause)
Parameter UCL
* *
* * *

* * *
* *
* * *
* *
* * LCL
* (Assignable Cause)

123 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Lot Number

The variations of Lot numbers falling within the UCL and LCL are Chance variations
which are acceptable and the process is under control

Variations of the lot numbers falling out side the UCL and LCL are Assignable
Variations the reasons for which are to be investigated to put the process under

3. R Chart (Range Chart)

Range Mean = R
LCL = D3 R
UCL = D4 R
D3 & D4 constants for sample size n for a certain confidence level (available in

* (Assignable cause)

R Mean

* (Assignable Cause)

123 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Lot Number

A quality inspector took 5 lots at 45 mts interval from a continuous glue
manufacturing process. Each time he took 4 samples and recorded Glue setting time
as below Lot No
Sample 1 2 3 4 5
1 12.11 12.15 12.09 12.12 12.09
2 12.10 12.12 12.09 12.10 12.14
3 12.11 12.10 12.11 12.08 12.13
4 12.08 12.11 12.15 12.10 12.12

Also given the following factors at 3 (taken from standard tables for A2, D3 and D4

Sample size n A2 D3 D4
3 1.02 0 2.57
4 0.73 0 2.28
5 0.58 0 2.11
6 0.48 0 2.00

Are the process under control at 3 control levels ?

1. Mean Chart
Sample size = 4
No of lots K = 5
x = Sample means
= 12.10, 12.12, 12.09, 12.10, 12.12
Eg. 12.10 = (12.11+12.10 +12.11+12.08)/4

x = x / K = 12.11

x = Sample mean
R = sample range = Difference between the maximum and minimum values is one lot
or Sample
Ranges = 0.03, 0.05, 0.06, 0.04, 0.05
R = Average of sample ranges
= R / K = 0.046

A2 (for n=4) = 0.73

UCL and LCL of x chart = x A2 R

UCL = 12.11 + 0.73x0.046 = 12.144

LCL = 12.076

R Chart
D3 = 0, D4 = 2.28
Range Mean = R = 0.046
LCL = D3 R = 0
UCL = D4 R = 0.105
All values are with in UCL and LCL in both the charts
Hence the process is under control with 99.75 % confidence level.

Process Control Charts for Attributes

Attributes are product qualities that cannot be represented by measurements like
dimensions, temperatures etc. but can be represented only by numbers.
The classification of each item is only as Confirming or Non-Confirming .
E.g. Good or Bad ; Pass or Fail,
e.g. 1. No. of wrong invoices vs no.of right invoices
e.g. 2. proportion of wrong invoices to total invoices made in a month.
e.g. 3. no. of fatal accidents vs non-fatal accidents in a year
e.g. 4. no. of painted part with < =3 defects vs >3 defects

Control charts for Attributes are

p chart
c chart and

p Chart
p chart is to monitor number of defectives generated in a process.
Formula for p chart
Sample size =n
Number of lots =K
Number of defectives in each lots = c1, c2, c3 etc

Fraction defective pi = c1/n, c2/n etc

Centre line (similar to mean) = p

p = K

UCL = p + 3 n

LCL = p - 3 n

Mean (center line) = n p = K

UCL = n p + 3 np(1-p)
UCL = n p - 3 np(1-p)

In a telephone company, an inspector took 100 Telephone bills each month from the
bills send to customers in the past 20 months. The number of defects identified in each
sample is as follows.
Sample 1 2 3 4 5 6
No. of defectives 4 10 12 3 9 11

Samples 7 8 9 10 11 12 13
Defectives 10 22 13 10 8 12 9

Samples 14 15 16 17 18 19 20
Defectives 10 21 10 8 12 10 18

Construct a process control chart that will describe 99.74 % of variations in the
process when it is under control.
Is the billing process under control?

Fraction defectives p = 0.04, 0.10 etc

No of samples K = 20
Sample size n = 100

p 2.2
Centre line p = K = 20 = 0.11

UCL = p + 3 n

= 0.11 + 3 100 = 0.20

LCL = p - 3 n
= 0.11 - 3 100 = 0.02


Samples 8 and 15 are beyond the control limits.

They will have some assignable reasons for the defects.
e.g. system malfunction, human error, unskilled entry operator.

They are to be investigated and corrected

c Chart

When the goal is to control the total number of defects in one unit or piece, we use c
chart to control the process

Eg. Number of defects in a hotel room. In a hotel room defects can be dust on TV, tap
leak, door very tight etc.

The goal set by the management is that the total number of defects in any room should
not be more than 3.

If they have say 100 rooms, it is not possible to inspect all the rooms.

So they chose random sampling

Management can stipulate that House keeping process should be under control at 3
sigma level (99.74 % confidence level). An inspector can prepare c chart, UCL and
LCL and certify the house keeping is under control

E.g.2. No of defects in an automobile, a printed page or Carton etc

Formula for c Chart
It assumes Poisson distribution and the probability of occurrence of more than same
defect in all the pieces is negligible.

Number of lots =K
Number of defects in each sample =c

Center line (Mean) = c

c = K

UCL = c + 3 c

LCL = c - 3 c


Lot number

Acceptance Sampling

Process Control Cart is used to ensure that the production process is under control
giving the required quality.

Where as Acceptance Sampling is for the parts that are purchased, or Finished
Products that go out for sale. Eg. 10000 bearings, 50000 Oil Seals

Acceptance Sampling assumes that the distribution of sample is similar to the

distribution of the population.
Need for Acceptance Sampling
1. In high volume production activities it is not possible to inspect 100 % items,
that are purchased

2. It will be expensive and time consuming

3. Acceptance Sampling is the only method when destructive testing is required

e.g. to find the strength of nuts and bolts.

4. Handling of part may cause damage a good part in 100 % inspection.

5. Prevent physical and mental fatigue to inspectors.

6. Need to inspect the parts more quickly to prevent inventory hold up during

Uses of Acceptance sampling

1. Decide acceptability of incoming items (OC Curve)

2. Decide the quality of out going products.(AOQ graph)

3. Also used for acceptance of group assemblies (e.g. gear box) in mass
production of automobiles

4. Use acceptance data in vendor rating

5. Widely used for material in the form of gas, liquid, stored in heaps, tanks or
moving in conveyor etc.

Types of Sampling Plans

Key element of Acceptance sampling is a Sampling Plan

A Sampling Plan stipulate

Lot size or population N
Sample size n
Number of samples taken - Single, Double or Multiple
Acceptance Number c , lot with defects beyond which are fully rejected.

Single Sampling Plan

One random sample (n nos) is taken from the lot (N nos). All n nos are inspected
100% and categorized as good or defective. If the number of defectives are <= c,
the lot is accepted, or else rejected.

Eg. 100 nos are taken from a supply of 2000 bearings and fully inpsected
If > 2 nos are defectives in the sample of 100, the whole of 2000 bearing are rejected

Double Sampling Plan

If offers an opportunity to take another sample if a lot is rejected in one sampling

Double sampling plan will specify two acceptance nos c1 and c2 for the initial
sample and single acceptance no c3 for combined samples
If the defectives are <= c1 , lot is accepted.
If the defectives are > c2, the lot is rejected
If the number of defectives falls between c1 and c2, a second sample is taken and
If the defectives of combined sample is <= c3 , lot is accepted or else rejected.

Eg. N= 2000, n = 50, c1=1, c2= 3, c3 = 2

Multiple Sampling

This is similar to Double sampling

The initial sample and all subsequent cumulative samples will have their own upper
and lower acceptance numbers. c11, c12 ; c21, c22 ; c31, c32 etc,
In the first sample
if the defectives are <= lower limit c11, the lot is accepted
defectives > upper limit c12, lot is rejected
If they fall in between c11 and c12, a 2nd sample is taken and inspected

In the combined sample

if the defectives are <= lower limit c21, the lot is accepted
defectives > upper limit c22, lot is rejected
If they fall in between c21 and c22, a 3rd sample is taken and inspected.
This process is continued when ever number of defects fall between upper and lower
limits and stopped when an accept or reject verdict is possible

Advantage of double and multiple sampling is that the sample size can be smaller
compared to single sampling resulting in saving in time and cost of inspection.

Multiple sampling is adopted for purchases made from reputed companies eg. SKF
Bearings, Intel Mother board, Intex keyboard etc

Choosing a Sampling Plan

- selection of the sampling plan depends upon the cost of inspection, cost of part
- if cost of inspection is low, single sampling with high sample size is used.
- if cost of inspection is high and/or cost of part is high, double or multiple
sampling is used with smaller sample size.

Operating Characteristics Curve (OC Curve)

OC Curves are used to do sampling inspection of Purchased Parts in order to decide

whether to accept the lot or not.

Important feature of any sampling plan is how to discriminate between good lot and
bad lot through sampling.

100 % inspection only can clearly discriminate between good and bad parts

In any sampling plan, some bad lots may get accepted and good lots may get rejected

The ability of a Sampling Plan (sample size n and acceptance no c) to

discriminate between good and bad lots with certain degree or probability of
accuracy is represented by the Operating Characteristics Curve of the Sampling Plan.
OC curve is a tool used by managers to decide the type of Sampling Plan to be applied
to different suppliers and supplies.

X Axis - Lot quality (fraction defective)
Y Axis - Probability of Accepting the lot

- Each OC curve is for a specific Sampling Plan with specific Sample Size (n) and a
specific Acceptance Number (c)

- With different combinations of n & c, the OC curve becomes more

discriminating ie. steeper or less discriminating ie. wider

- Any point on the OC curve will show the probability of accepting a lot with
certain fraction defectives in the whole lot.

Sampling Terms
1. Acceptance Quality Level (AQL)
AQL is the fraction defectives (percentage of defects) at which consumer (buyer)
is willing to accept lots as good

In the graph AQL is 0.03 %

2. Producers risk () Type 1 error

Producers risk (ie. suppliers risk) is the probability that a lot containing the
acceptable quality level will be rejected

In an OC curve, the AQL line meet the curve at certain point, which indicate the
probability of rejecting a good lot containing acceptable quality.

In the graph the producers risk is 10%

3, Lot Tolerance Percentage Defects (LTPD)

LTPD is the upper limit on the % fraction defects that a consumer (buyer) is willing
to accept

In the graph, LTPD is 0.1%

4. Consumers risk () Type 2 error

It is indicated by the point where LTPD line meets the OC curve

Consumers (buyers) risk is the probability that a lot containing defectives exceeding
the LTPD will be accepted.

It means that the end products of the buyer will contain some defective parts which
will be passed on to the end consumer with that probability

In the graph Consumers risk is 10 %

In effect, the manager and supplier have to sit together and agree for a Sampling Plan
(n &c) and OC curve in line with AQL and LTPD stipulated by the company so that
both Producers and Customers risks are at acceptable levels for both.
Ideal OC Curve (100 % inspection)
Ideal OC curve for
100 % 100 % inspection

probability Reject
of acceptance Accept

0.005 0.01 0.015 0.02

Fraction defectives (%)

In 100 % inspection of the lot, the probability of accepting lots with defects =<0.01%
defect is 100 % and >0.01% is nil.
This is because the full lot with defect =< 0.01% is accepted, if more > 0.01%, the lot
is rejected

Average Outgoing Quality (AOQ)

AOQ curve is used to ensure that the quality of out going finished product is with in
the Average Quality specified by the company before they are sent to distributors.

Statistically an AOQ curve is as given below

Outgoing Quality Level (OQL)
defective 0.06


0.05 0.1 0.15 0.2 0.25
In coming Fraction defective

It works the flowing way

- the company fixes the AOQ level as 0.08 %
- Samples are taken from the incoming goods from the production line.
- When the lots are below 0.08 % defects, they are accepted
- When a lot is having more defects, say 0.15 %, this lot is subjected to 100 %
inspection and % defect for this lot becomes 0.
- More attention is given to production process by rejection analysis of this lot
- By this method, the overall average out going quality can be kept below 0.08 %

Six Sigma Approach to Quality

In statistical terms, Six Sigma quality means less than 3.4 defects
per million opportunities for defect in any process, products or
services ( compared to 3 sigma, ie, 2.4 defects in 1000 Nos)


-6 +6
Quality parameter
(dia of a shaft)

In Operations Management, Six Sigma has much wider meaning.

It is a program to achieve
- zero defect
- improved customer satisfaction at
- lower cost and in lesser time

Six Sigma is essential for global companies, that manufacture goods

for world market. For them, 3 sigma limit is totally inadequate to
meet customer satisfaction and competitiveness.
Eg. HP produces 3 lap tops per minute in their US plant and annual
production is about 9 lakhs
At 3 sigma quality (99.76 % confidence level) in the final assembly
alone, the number of defective laptops per year is 2160.
If one takes cumulative defects from 3 sigma quality at component
level, the defective laptops will several thousands and it is not
acceptable to HP.
Hence they go for Sigma quality level i.e only 3 defective lap tops
per million produced

Components of Six Sigma

Six Sigma has both management component and technical
Management components
- provide strong leadership and commitment
- define performance matrices

- select projects that are likely to achieve the objective.

- Select and train people.

Technical components are

- improve process capability of machines, methods, maintenance
and inputs to 6 sigma level

- reduce variations in the out put

- use statistical methods for variation detection, analysis and


- designing a structured continuous improvement strategy which

include definition, measurement, analysis, improvement and

Key players in Six Sigma Program

- Top management formulate and communicate overall Six
Sigma Program and lead the program.

- Champion Six Sigma Leader, who identify and rank potential

projects, help select and evaluate candidates, manage program
resources, serve as advocate for the program

- Master Black Belts They are specialists in Six Sigma Program

having extensive training in statistics and use of quality tools.
They train Black belts

- Black Belts they are project team leaders responsible for

implementation of process improvement project after getting
trained in six sigma program and demonstrated mastery in the

- Green belts are project team members.

Typical six sigma programs are to

reduce defects,
reduce cost,
reduce product or process variability,
reduce time,
increase productivity
and improve customer satisfaction.
Motorola was the first company to pioneer in Six sigma in 1980s.