TQM TME016 uptu

8/2/2019 TQM TME016 uptu

1/16

AKGEC/IAP/FM/01

REV. No. 01

AJAY KUMAR GARG ENGINEERING COLLEGE

GHAZIABAD

UNIVERSITY PAPER SOLUTION

Course : B.Tech. Name of Faculty : Narendra Kumar

Semester : VII Session : 2010-2011

Subject Code : TME-016 Subject Name : TQM

Q.1.

(a) How has the concept of Quality evolved with time, discuss various phases of Quality control

along with their characteristics.

Ans- Quality evolution started from the stone age the user and the producer of the tool were the sameperson. They modified the sharpening of the tool after actual performing the tasks so in those days

quality was check by user itself. Important steps in the quality evolution are-

Craftmanship From early civilization to the industrial revolution, quality control was exercised by thecraftman himself, or the master craftman who guided a group of craftmen. The reputation for

craftsmanship and the incentive of better price for a superior product were the main governing factors of

product quality.

Supervisors Control With the expansion of manufacture during the industrial revolution, large

factories were established to meet the increasing demands of the consumers. This give rise to the need ofsupervisors to control the work of a large number of workers. Usually, supervisors were picked up from

among the elder workers who had thorough knowledge of the job, and thus were able to guide andadminister the other workers. Among other functions, the checking of the quality of jobs produced in theshop was also the responsibility of the supervisor or the foreman. This was usually by visual check of

jobs, for which he mainly relied on his subjective experience.

Inspection With the advent of the manufacture of interchangeable components on a mass scale in the

early part of this century it became essential to critically examine each component with the aid of

measuring instruments or gauges. Obviously this job could not be performed by the shop supervisor,

who was fully occupied with other problems concerning production. Thus a full-time inspector came onthe scene and the quality function was separated from production. With these developments a new

discipline, Engineering Inspection, came into being and elaborate inspection organization were created

in large manufacturing companies.

Statistical Quality Control World War II generated a great upsurge in industrial activity. The

tremendous requirements of defence stores necessitated mass production on an unprecedented scale. Asconventional inspection methods were inadequate for this purpose, statistical methods like sampling

tables and control charts were introduced, which reduced the burden on the inspection organization, and

enabled it to cope with mass production. Although, statistical quality control to include in-process check,

it still remained an extension of inspection, as it tackled the quality problem only at the shop floor levelby the physical inspection of a part or whole of the product. It could not view the problem from the

overall company perspective as the management saw it.


2/16

(b) How does quality of design differ from quality of conformance?

Ans- Design is the basic function for the creation of product quality. Unless quality is designed into a

product, it cant be achieved during manufacturing. The prime aim of designer is to create a product that

will satisfy a customers need at a price which is willing to pay. With the introduction of a newer andbetter product as a result of technological progress. The customer is becoming increasingly well

informed and discriminating for superior quality claim made by aggressive salesman in competitive

market, the design have to keep on striving for higher and higher quality level, if their companies are toremain in business. So, we can say a quality product cant be produced without quality design in the

product.

Quality of conformance relates to the fidelity with which like product confirm to design. It requiresattention to appropriation of resources like technology, manpower and managerial processes which are

used for translating the design into acceptance products. Degree of conformance 100% implies the

product has been produced as per design quality of required of product zero defect. Zero degree ofconformance means, the product is totally defective.

(c) What is service quality? What are internal and external measures of service quality? (NA)

(d) Draw and discuss a typical Process control network used to ensure quality of end product.

Ans-The process controls discussed so far, mainly pertain to the activities which are required to be

carried out at any particular stage of the process. Often, the final product will comprise of a numberof components and assemblies, some of which may be purchased items and others may be processed

in the same plant where the final product is assembled. To ensure the satisfactory quality of the end

product, quality control should cover the various stages of manufacture of all components andassemblies which are processed in the plant. In addition, it should also cover the quality assurance of

all incoming materials and components. This will require a network of control stations located at

strategic points in the plant. At each control stations, the quality characteristics to be controlled, should beclearly defined, and adequate test equipment and trained staff should be provided to carry out this

function. Detailed instructions covering the integrated quality control network have to be issued to

ensure the smooth working and success of the quality control programme. These instructions shouldinclude:-

(a) Quality characteristics to be controlled.

(b) Delegation of responsibility for the control station.

(c) Procedure for the drawing of samples test pieces and methods of checking/testing of the

quality parameters.

(d) Procedure for the analysis of inspection data, including criteria for taking decisions regarding,

acceptance of process/product quality:

(e) Procedure for transmission of information about the process quality to the production point

and responsibility for carrying out process adjustment.

(f) Procedure for the feed-back of information to the process planning and quality planningdepartments.

To plan a process control network, it is first necessary to have a flow process chart for all the

components and assemblies. From this chart, a note is made of critical points where imporz tant

quality characteristics are generated. Then, depending upon the nature of the process and the qualitycharacteristic of the product, a decision is taken, regarding the control technique to be employed at

each of the critical points. The exact location of the control station will depend upon the control

technique, location of production points and availability of the test equipment. For example, if aprticular quality parameter can be easily checked without elaborate test equipment it may be decided to

use control chart mounted on the processing equipment itself and the operator himself or his

supervisor may be asked to take measurements and maintain the control chart. On the other hand,


3/16

the inspection of a component may require considerable time, and expensive test equipment and there

may be a number of machines producing the same component. In such a case one control station,located at a convenient place, may be established, where components produced by all the machines

may be brought for post-process inspection. To economise on the deployment of quality control

staff, common control stations may be established for controlling the quality of a number ofcomponents. Alternatively, one quality inspector may be made responsible for a more than one control

station and thus act as a roving quality controller. Control stations should be so organised that they

fall in the normal flow of the process and require a minimum of handling and movement of stores.

Particular care should be taken to see that the control station do not create bottle-necks in the processand the capacity of an inspection station is commensurate with the production rate of the process. A

process control network is illustrated in Fig

given below-

Q.2.

(a) What are the various categories of cost of Quality and their constituents?

Ans -In Quality Control Practice any cost which occurs to ensure that the outgoing product is of

requisite quality is termed as Quality Cost. Quality Cost goes on increasing for better

Quality Conformance.Quality Costs can be considered under the following three headings-

Failure Cost

i. Internal failure cost such as cost of scrap and rectification and reduction in sale price of 2nd

quality grade.

ii. External failure cost such as replacement during warranty period, expenses on

investigation and adjustment of customer complaints.

Appraisal Cost : These represent cost of inspection and testing during various stages of

manufacturing as well as inspection of incoming material.

Preventive Cost : These include cost associated with activities such as quality planning


4/16

process control and quality training which are aimed at preventing defective jobs being

produced.In Competitive induction whose quality of a product is of great importance.

Optimization of Quality Cost:- As the Quality of conformance increases the quality cost vary as shownbelow

The Total Quality cost comprises of the three elements in varying proportion as shown below

At quality level 1X , the total cost is 1Yof which failure cost 7.5% and balance 2.5% is made of preventive &

appraisal cost.

At quality level 2X , the total cost is 2Y which is made less than 1Y and at the level failure costs are 50% of

total quality cost.

At quality level 3X , the total quality cost is 3Y which is again higher than 2Ybut is mainly comprises of

prevention costs. Thus, it can be seen that as quality of conformance increases, the quality cost at first

decreases and reach a certain minimum and then again begin to increase. Therefore from the economic of


5/16

view, quality level 2X where Total Quality cost is minimum, represents the optimum level of the quality of

conformance. In practice, a little higher quality of conformance is taken then optimum level to enhance the

reputation of organization.

(b) Using a schematic diagram, explain the various steps in the construction of the QFD house of

quality. (NA)

Ans- The term quality function deployment that represents the overall concept that provides a meansof translating customer requirements into the appropriate technical requirements for each stage of the

product development and production. The customers' requirements expressed in their own terms areappropriately called the voice of the customer. Technical features are the translation of the voice

of the customer into technical language. They are the "hows" that determine the means by which

customer attributes are met. A set of matrices is used to relate the voice of the customer totechnical features and production planning and control requirements. The basic planning

document is called the customer requirement planning matrix. Due to its structure it is often

called as the House of Quality. The House of Quali ty relates customer attributes to technicalfeatures to ensure that any engineering decision has a basis in meeting a customer need. A simple

example of a House of Quality for the hypothetical case of a quick service franchise that wishes to

improve its hamburger could explain the process involved in evaluating customer requirement. Thevoice of the customer includes four attributes.

The hamburger should:

Be tasty Be healthy

Be visually appealing, and

Provide good value.The technical features that can be designed into the product are price, size, calories,

sodium content, and fat content. The taste bears a strong relationship to sodium content, a

moderate relationship to fat content, and a weak relationship to calorie content. The competitiveevaluation would show that competitors are currently weak on nutrition and value; these can become

key selling points in a marketing plan if the franchise can capitalize on them. At the bottom of

the house, are targets for technical features based on an analysis of customer importance ratings and


6/16

competitive ratings.

The House of Quality provides marketing with an important tool to understand customerneeds and gives top management strategic direction. However, it is only the first stage in QFD

process. The voice of the customer must be carried throughout the production process. Three other

Houses of Quality are used to deploy the voice of the customer to component parts,characteristics, process planning and production planning. These are:

Technical features deployment matrix: Ittranslates technical features of the final product into

design requirements for critical components.

Process plan and quality control charts: It translates component featuresinto critical process and product parameters and control points for each.

Operating instructions: It identifies operations to be performed by plant personnel to ensure

that important process and product parameters are achieved.Most of the QFD activities represented by the first two Houses of Quality are performed by people in

the product development and engineering functions. At the next stage, the planning activities begin to

involve supervisors and production-line operators. This represents the transition from planning toexecution. If a product component parameter is critical and is created or affected during the

process , it becomes a control point. Th is te ll s the company what to monitor an d in spect and

forms the basis for a quality control plan for achieving those critical characteristics that arecrucial to achieving customer satisfaction. The last house relates the control points to specific

requirements for quality assurance activity. This includes specifying control methods, sample sizesand so on, to achieve the necessary level of quality.

The majority of QFD applications in the United States concentrate on the first, and to someextent the second, House of Quality. The third and fourth houses offer far more significant

benef it s, especi al ly in the United States. Japanese managers, engineers and workers are more

naturally cross-functional and tend to promote group effort and consensus thinking. US workersare more vertically oriented and tend to sub-optimize for individual and/or departmental

achievements. Beginning to emphasize effective cross-functional interactions as supported by

QFD will enable U.S. firms to be more competitive with foreign rivals. The third and fourthhouses of quality utilize the knowledge of about 80 percent of a company's employeesif they

are not used, this potential is wasted. 09

Building a House of Quality requires six basic steps:.1. Identify customer attributes

2. Identify technical features

3. Relate the customer attributes to the technical features of a design.

4. Evaluate competing products based on customer attributes.5. Evaluate technical features of a design and develop targets.

6. Determine which technical features to deploy in the production process.

Steps in a QFD process can be split into following sequential measures:

1. Identify & Establish Customer Attributes. The first step is to identify and establish customer

attributes. Care has to be taken not to allow any distortion or misunderstanding of the

attributes wanted by customers during the processing of information. If there are more thanone customer in the chain, then all the needs should be considered.2. Technical Features Necessary for Customer Satisfaction. The second step is working out and

listing all technical features that are necessary for meeting the customer satisfaction. Thesetechnical features should form the basis of activities for subsequent design, manufacturing and

service processes. The technical features must be quantitative and measurable in order to

facilitate control and comparison to target values.3. Establish a Relationship Matrix. The third step is to establish a relationship matrix between

the customer attributes and technical features. Purpose of this relationship matrix is to

ensure that final technical features adequately address all the customer attributes. Often,this is established by controlled experiment or getting feed back from customers. Lack of


7/16

strong relationship. between customer attributes and any of the technical features may

lead to failure of meeting the customer needs.1. Rate Area of Interest for Customer. This is the step for rating customer importance about the

areas of greatest interest and highest expectations. Taking care of this in the des ign fea tures

adds market value and can act as key selling points. In this regard, evaluating competingproduct or service become very usefu l. By fo llowing th is st ep of QFD method, product

development process can be linked to company's strategic vision and goals.5. Develop Target Quality Features. The fifth step is the evaluation of technical fea tures of

competitive products, review of own technical features and development of `targetquality' features. This can be done by in-house testing or reverse engineering, and

translating the results into measurable terms. Then, targets for each technical feature

are set on the basis of customer importance. At t imes, though the technical featureanalysis may show otherwise, a competitive product may still enjoy high customer

sat isfaction rating due to some perceived view.5. Identify Product Characteristics & Customer Needs. The final step is identifying the

features and product characteristics that have a strong relationship to customers needs

and have strong selling, points. These characteristics have to be then deployed in the

design and producti on processes, actions and measures for control have to be put intoposi tion for ensuring the product or service attributes as per the original 'voice of

customers'.These 'customer requirement planning matrix' can be placed in chart depicting a structure

called the 'House of Quality ' shown in Fig. 8.3. House of quality provides critical understandingof customer needs and also gives strategic direction to leadership.

(c) What are the causes of operators errors? Discuss the corrective measures.

Ans- Operator's errors result from three main causative factors, namely,

(a) Incompetence

(b) Lack of awareness

(c) Carelessness and lack of interest.Incompetence

Operator can be expected to achieve quality conformance only if he has the necessary skill. Some operators

may be able to achieve the quality targets consistently, while others may not be able to maintain the required

quality level inspite of their best efforts. The difference may be the skill. Once this deficiency has beenidentified the remedy lies in training the operators to enable them to acquire necessary proficiency in their

job.

Lack of Awareness

There are some errors which are made by the operators without their being aware of it. Such errors can be

minimised by making the process as fool-proof as possible. Use of fixtures which automatically ensure thatthe job is correctly positioned for the operation, or an assembly design where the product cannot be

assembled the wrong way, are some of the examples of fool-proofing. In addition, certain devices may be

incorporated in the process, which provide a visual or sonic alarm when an error is made, so that theoperator is alerted in time Other measures to reduce such errors are proper lighting, clean working

conditions and avoiding undue fatigue and long working hours.

Carelessness and Lack of Interest

The majority of the operator's errors are caused by this factor. How has this situation been broughtabout, where the operator or worker is apparently not interested in the product he manufactures? The

main cause of his disinterest is the lack of knowledge about the product. Seeming indifference on the


8/16

part of operators to the quality of the product is not due to the fact that they want to produce bad jobs.

It is mainly due to ignorance and lack of understanding of the implications of poor quality. The oldcraftsmen made all the components of a product themselves, and saw the result of their labour in the

final product. Thereby they were aware of the flaws if any in the finished product, and took pains to

improve the quality by eliminating the flaws in the next piece.

In the modern mass production technology, the product has been broken down into small discrete

elements, and a worker may be producing thousands of small parts every day, without knowing wherethey fit, what is their function and what may happen if they are not made exactly according to the

specification. Under these circumstances he cannot be blamed if he is not very much concerned about its

quality.

The corrective measures are 1. Motivation of Workers

2. Education of the Workers

3. Financial Incentives4. Workers Involvement in Quality Improvement

Q.3.

(a) Explain the difference between attribute control charts and variable control charts.

Ans - Measurement Charts X-Chart & R-Chart we take measurement with instrument for sizethen range & Mean charts are drawn for samples to be taken. Each sample is consisting of

few subgroups.

X-Chart

X = N

XN = No of Groups, n = No of subgroup in a sample

UCL = X + 2A R

LCL = X - 2A R

R-Chart

UCL = 4D R 3D , 4D , 2A , 2d for n subgroup in sample are taken from Table.

LCL = 3D R

Attribute charts: In this type of control chart we need not to take measurement only visual

inspection is done and find defective units under total units inspected or no. of defects in each units is

find out Caverage defect/unit is calculated

P - Chart Fraction defective chart

CP - % defective chart

P- Chart

P =


9/16

UCL = P + 3 N = No of subgroup in a sample

LCL = P - 3

CP -Chart % defective chart

CP = P100 =

UCL = CP + 3

LCL = CP - 3

C-Chart defect chart

C =

UCL = C + 3 C

LCL = C - 3 C

C-Chart Weighted control chart

Demerit points are given to each type of defect

C =

UCL = C + 3 C

LCL = C - 3 C

(b) Under what circumstances should c-charts be used instead of p-charts?

Ans- P-charts are based on defective bodies. If one or more defects are there, the body will be defective. Weuse P-charts. But when no of defects are more the body may be accepted or rejected depending upon

no. of defect per body to be acceptable looking into their effect on performance than we use c-chart i.e.defective chart or weighted chart or demerit chart are based on demerit points of each type of defect

depending upon severity or criticalness of each type of defect. Example Cast Iron engine body having

no. of below holes, so we have to take decision how much defects are acceptable with out effecting theperformance.

(c) What are the natural limits of a process? How do they relate to product specification limits?


10/16

Ans- Natural limit of a process are its process capability which is given by 16 i.e.2

6d

R

Where, 1 is the estimated standard deviation for individual for individual observations.

R is the average range of sample ranges.

2d is afactor which varies with sample.

Product specification limits are the maximum & minimum size of any product, the difference of

maximum size & minimum size i.e. Tolerance for no rejection process capability must lie withintolerance.

Example IfR =0.035 mm, n=5, 2d =2.326

Process capability =326.2

035.06

= 0.090 mm

Specific tolerance for Pin 0.06 mm

Tolerance range = 0.12 mm

Thus it cab seen that process capability of process is well within the tolerance range of the

product. [0.09 0.12]

(d) Explain the different information provided by the process capability ratio and the process

capability index.

Ans- Process capability is the quality performance capability of a process under in-control conditions.

After X and R control charts show that the process is in control, process capability study is generally

undertaken to find out whether the process can meet the job tolerance limits. It will be recalled that

this aspect is also checked earlier during analysis of frequency distribution before installing controlcharts. The purpose of this recheck is to substantiate the findings of earlier analysis. It should not

be forgotten that the earlier frequency distribution was studied when the process was newly set-upand it may not have settled down. Further, over a time certain factors may have altered the

variability of the process, therefore, it is always a sound practice to study process capability before

commencing a long run of a product.

Mathematically process capability is defined as 6 standard deviations. It should be noted that this

standard deviation pertains to the individual observations on items or pieces, which can be obtainedfrom the mean range (R) by using the following relationship.

2

1d

R=

Where, 1 is the estimated standard deviation for individual for individual observations.

R is the average range of sample ranges.

2d is afactor which varies with sample.

(e) Enumerate the advantages and disadvantages of cusum charts. (NA)

(f) What is average run length (ARL)? (NA)


11/16

Q.4.

(a) What is failure rate for reliability 0.95 and hours 400? What will be reliability after 100 hours?

Ans-tetR =)( where R(t) is reliability at time t.

is failure value.

400)400(

= eR

te =95.0

40005129.0 =

4102823.1 = per hour

Reliability at 500 hr (i.e. after 100 hour)

500)500( = eR

500102823.14

= e

064116.0= e

Reliability after 100 hours, 9378.0)500( =R

(b) Explain how to compute the reliability of series and parallel systems.

Ans-Reliability of Parts in Series

Suppose a system consists of a number of parts, each of which is vital for the functioning of the system, inthe sense that the failure of one part results in the failure of the whole system. The parts in such a systemcan be considered to be in series as shown in Fig.

If the reliability of each part be 1r, 2r , 3r the reliability ( s ) of the whole system is given by :-

Let us consider the example of a warning system, which consists of switch a sensor unit and a lamp.

For simplicity's sake let us assume that each of the three parts has a reliability of 0.9. The reliability of

the whole warning system is given by,

sR = 0.9 x 0.9 x 0.9

= 0.72.

Now let us suppose that a system comprises of 20 parts each with 0.9 reliability. The reliability of such a

system will be,

sR = (209.0 )


12/16

= 0.12

0.12 Reliability means that there are only 12% chances that the system would work, which is a

statistical way of saying that the system is useless. Now a system with 20 parts is very simple according to

the modern standards. A radio or domestic refrigerator has more than 100 vital parts, a car, morethan 1000 and a large aircraft over a million. How can a reasonable level of reliability for such

complex systems be achieved? Firstly, every effort must be made to reduce the number of parts in the

system. To achieve this end, the value Analysis technique discussed in chapter 18, have been found to

be very useful. Secondly, we should have components with as high reliability as possible. In the above

example we had assumed component reliability to be 0.9. Luckily, with the technological progressmade in the last few decades it is now possible to achieve much higher level of reliability and

components with 0.999 reliability have become quite common. Let us study the reliability of a systemusing parts with a reliability of 0.999. Suppose the system has 400 parts in series, its reliability

will be ( ) 400999.0 or 0.673. Now a system where the chances of successful operation are 67% or 2 outof 3, cannot be called very reliable. Therefore, in addition to using components with high individual

reliability, certain other measures are also required to improve system reliability. Some of these arediscussed in the following paragraphs.

Suppose a particular component in the system is found to be unreliable and it is decided to use

two similar elements in parallel so that if one fails, the other can carry on the function. Let the twoelements be designated as No. I and No. 2 as shown in Fig. (21:3) and their reliability be R, and R2respectively. The system will fail only if both the elements fail, the probability of which can be

calculated as follows:

Probability (F1) that element No. 1 fails = 1 - 1R

Probability (F2) that element No, 2 fails = 1 2R

Probability that both No. 1 and No. 2 fail = 21 FF

The system will continue to work satisfactorily until both the elements fail. Thereforereliability ( )cR of the Combined unit of two elements is given by

( ) ( )211 FFRc =

The same equation can be written in the following general form when there are K elements in

parallel.

( ) ( )kc FFFFR ..........1 321 =

)......().........1.().........1)(1)(1(1 321 iiRRRR k=

For example, if 3 elements are in parallel and each has a reliability to 0.80, the

Lool

Calc

I

cu mb


13/16

reliability cR of the

combined unit is given by,

R = 1 - (1 0.8) (1 0.8) (1 0.8)

= 1 - 0.2 x 0.2 x 0.2

= 1-0.008

= 0.992.

From the above it can be seen that theoretically, we can raise the reliability to any desired level by

Means of redundancy, provided we can afford to have additional spare items. Practically, however,there is a limit to the use of redundancy. Besides the expense incurred on additional elements,

redundancy adds to the bulk and weight which may cause other problems. Secondly active

redundancy may require sensing and switching devices, which bring in additional sources ofunreliability. Therefore, redundancy is attempted only as a last resort.

(c) Discuss technique for reliability analysis.

Ans- Unlike other quality parameters, the reliability of a product cannot be directly measured or checked. At

best, an estimate of the reliability of the product lot can be made by testing a few pieces, as is done in the

case of sampling inspection. Reliability testing generally involves running the product units under testfor the duration of their expected life. The proportion of the units which fail during the test gives an

indication of reliability of the product. The product units subjected to reliability test have to be written

off, as at the end of test either the units would have failed or be in the last stage of their useful life. Incase the product is expensive the cost of test pieces may place a limit on the number of pieces which

can be tested. Further, the life of a product may be thousands of hours or even more. To test the

product for this duration may be impracticable due to the limited time available. This problem may beovercome by adopting one of the following approaches.

(a) Accelerated Testing A device may operate intermittently in actual life. Testing time can be reducedby operating the device more or less continuously. In such cases, care must be taken to see that

continuous operation does not lead to undue rise in temperature which may cause premature failureof the device. While testing devices which are subject to this tendency, adequate rest periods must be

allowed for the device to cool down.

(b) Test of Increased Severity Sometimes, it may be possible to carry out the test under more severe

conditions than are met in actual service. This may be achieved by increasing the loading andimposing harsher environmental condition such as temperature, pressure, vibrations, etc. If the

device does not fail during a relatively brief exposure to these conditions, it may be possible to

predict the life by extrapolations. However, this approach can only be adopted if a proper correlationhas previously been established between the actual observed life and the life under specified severe

conditions.

(c) Testing a Larger Number of pieces We have seen earlier that the failure rate during the usefulperiod of life of a product is generally constant. Instead of testing a few pieces for the entire period

of specified life, a larger number can be tested for a comparatively shorter time, until a few failures

have occurred. The failure rate can then be determined from the total hours logged by all piecesunder test.

(d) Write a note on maintainability analysis.

Next


14/16

Ans - It is the ability of any system to put into original efficiency within a given period.

If the time increases maintainability increases where reliability decreases.Different elements of MTTR:

1. Reporting of failure of system or production.

2. To send portable system to repair centre or to send repair team at site.3. Inspection of fault (fault checking).

4. To arrange spare for repair.

5. To change or repair the faulty parts.

6. To check the effectiveness of repair or replaced components after putting system in function &testing 3rd, 5th and 6th are essential components of MTTR .

Q.5.

(a) What is Taguchis loss function? Discuss.

Ans- Noted Japanese Quality philosopher, Taguchi, has developed a unique technique for reducing the

inherent variability in a product or process .; This is popularly known as Taguchi method of

'Quality Loss Function', which is a graphic representation of the loss to society caused by product

or process variations. The Taguchi philosophy is based on the premise that improving qualitycan reduce cost, and quality gets improved by reducing variations.

Taguchi method is a combination of engineering and statistical approach to problem solving,

and is primarily applied to product or process engineering to identify and optimize input factors

for improved quality at reduced cost. He described the factors in product and process

engineering as controllable and uncontrollable These factors interact during the design and

manufacturing, impacting the variations inherent in the product. Taguchi suggested that by

focusing only on the controllable factors and their resultant effect on variation, product could

be designed with minimum variation, leading to improved quality at reduced cost.

Taguchi's 'Quality Loss Function' graph is a parabola in which the lowest point represents the

minimum loss to the customers and company. This minimum point , corresponds to the 'target',

and any variation from target value in either direction is the cost to society. Increase of variationfrom target in either direction increases the cost, vide Fig.

Quality loss function analysis should be used during the early stages of design, so that

changes can be made as early as possible. This technique can also help in detecting if the


15/16

product is over designed. The method proposes to reduce the effect of noise factors on

variation by concent rating on the controllable factors, because those are the only factors that

can be changed for improvement.

Attempt to remove or control the uncontrollable (noise) factors makes the process and

product too expensive. Taguch i recommended use of 'orthogonal ar rays ' as a technique

to simulate the results of various factor combinations to reduce the number of experiments

otherwise necessary to complete the design.Continuous improvement is a creative process. It requires open mind and the belief

that improvement is always possible and there are several means of possibility of reaching

to that goal. Improvement is the process of choosing the appropriate one, and then

working out the best solution using that means i.e. the tool. Different tools discussed

here are by no means complete treatment of the subject; they have been discussed here

for giving a brief overview of their purpose and features. Details of some of the se too ls

can be f ound from the corre spond ing books listed in t he bibliography.

Purpose of this chapter has been to highlight the usage of tools and techniques for quality

planning and continuous improvement in TQM process. Stat is tical tools, which are

extensively used both in manufacturing and service sectors, have become indispensable as

support tools for quality planning and continuous improvement, as well as for other

business applicat ions . It would, therefore, be just to di scuss these tools separately in the

next chapter. Other techniques, which have been mentioned here but not discussed, are the

benchmarking, business process re-engineering and measure of cost of quali ty. These

techniques have the potential to bring about quantum benefits to the organization in terms of

performance.

(b) What is ISO? What is ISO 9000 quality system?Ans -ISO 9001 Quality Assurance model for design/developing production/Installation

and Servicing.ISO 9002 Quality Assurance model for production and Installation.

ISO 9003 Quality Assurance model for Final inspection and testing.

ISO 9001 It gives a model for quality assurance at all stages starting from designing the product and

continuing even after the product is delivered to the customer. It applies to industries that design,

produce, install product and provide service after sales as per requirement. It enables the customer to

judge the supplier capability of manufacturing the product as per his requirement.

ISO 9002 Assurance during production model for such product. In such cases the manufacturersgives his own design to meet the customer requirements and has only prove the production process iscapable of producing the product ex civil structures, construction of bridge etc. Assurance applied only

on production.

ISO 9003 Certain product require Quality Assurance only after they are magnified at the time of

supply. The consumer is only interested in getting the product of desired quality as stated by supplier

ISO 9003 gives model in this case.


16/16

(c) Explain the rules of switching between normal, tightened, and reduced inspection for acceptance

sampling by ISO 28591:1999. (NA)

(d) Discuss types of Quality Audits.

Ans-Quality audit is an appraisal of the whole quality control. It is not so much concerned with thequality of the product, as with the adequacy and effectiveness of the quality control system.

Quality audit may be conducted periodically, or only when occasion demands, due to existance of

quality problems. Quality audit includes the examination of the following aspects:(a) Completeness and clarity of the manufacturing drawings and specifications and procedure

for

their updating.

(b) Process capability of manufacturing equipment and adequacy of process controls.

(c) Quality control of incoming materials and procedure of vendor's capacity verification.(d) Adequacy and accuracy of gauges and test equipment and procedure for their calibration.

(e) Quality organisation and qt4ality control procedures.

(a) Reject rate of the product as revealed by in-process and final inspection results.

In addition to the above, samples of the important products may be taken for critical inspection in the

laboratory. The number and type of defects in the samples will provide an index of the effectiveness of

the control system. To be effective, audit must be independent of the established inspection and processcontrols. Audit results should be properly documented and forwarded to the Quality Manager as well as to

the concerned divisions and the sections of the company. Any discrepancies revealed in the audit should

be rectified within a reasonable period. Follow-up action can be progressed by the Quality Manager under

whose aegis quality audit is normally carried out. A word of caution here is, that Quality audit should be

regarded as a tool to help in improving the quality of a product and not for witch-hunting and punitive

action. Quality audit requires the active cooperation of all departments and sections concerned with the

quality of the product. This cooperation may not be forthcoming if there is fear from quality audit.

Documents

TQM TME016 uptu