用多属性的决定模型来评价全面质量管理的因素

TOTALQUALITYMANAGEMENT,VOL.13,NO.6, 2002, 779 -796


Multi-attribute decision model for assessing
components of totalquality management

R. S. Dalu
1& S.G.Deshmukh
2
1Mechanical Engineering Department, GovtCollege ofEngineering, Amravati,(M.S.) India&
2MechanicalEngineering Department, Indian InstituteofTechnology,Delhi,HauzKhas,New
Delhi, 110016, India

Continuous improvement is an imperative for being competitive. The studies show that
totalquality management(TQM) isanet ectivetoolforcontinuousimprovement.The literatureon
quality management shows that there is need of a model which can compare the industries on the
basis of components of TQM. This paper describes the multi-attribute decision model using the
analytichierarchyprocessfortheassessmentof the quality statusofindustry.

abstract

Introduction

Theliteratureon totalquality management (TQM) includesvariousTQM models (e.g.
Joiner triangle(Joiner& Scholtes,1985), Juran trilogy(Juran, 1989),Four-sided pyramid
model, House of TQM model, Malcolm Baldrige National Quality Award (MBNQA), the
Australian Quality Award (AQA), European Quality Award (EQA), etc.). All these models
are largely descriptive, with little or no emphasis on logical links between the components.
For the purpose of researching the value of TQM, these descriptive models are of limited
use (Saunders & Preston, 1994). The other limitation of these models is that they do not
makecleartheoverallbene.ttobe obtainedfromTQM,nordothey specifythebene. ciaries.

In 1994,SaundersandPrestondevelopedthe `S-P model ofTQM’.Thismodelprovides
auseful frameworkforstudy ofthe applicabilityofTQMindit erent economicenvironments.
It can also be used to identify the aspects of the environment required for successful
implementation of TQM.

ThelimitationsoftheS-P model arethatitcannot(i) identifythedegreetowhichthe
various components of the TQM are present in the organization; and (ii) compare the
industries on the basis of components of TQM.

Onthis background,theobjectivesofthis paperareto:(i) studythe variouscomponents
oftheS-P model;(ii) identify criteria and quanti. ableindicatorsfor measuring thecomponentsoftheS-
P model; and(iii)develop a multi-attributesdecisionmodel analytic hierarchy
process (AHP) to overcome the limitation of the S-P model.

The S-P model

The salientfeatureofthe model are as follows. They:(i)provideastructurefor research into
TQM;(ii) identify the logical links between the components of TQM and thus allow

Correspondence :S.G.Deshmukh,Mechanical EngineeringDepartment, IndianInstituteofTechnology,Delhi,
New Delhi, 110016, India.

ISSN0954-4127print/ISSN1360-0613online/02/060779-18 .2002 Taylor & Francis Ltd
DOI:10.1080/0954412022000010136


780 R. S. DALU & S. G. DESHMUKH


Figure 1. The S-P model.

identi. cation of those components necessary for successful application of TQM;(iii)provide
a basisfordetermining

howwidelyTQMis applicable;(iv) assessthe applicability ofTQM
to dit erent types of industries (small/medium/large), cultures and economic environment;

(v) help tostudy theet ectofexternal andinternalfactorsonthe componentsofTQM;(vi)
identify thepotentialbene.tstobeobtained through changesintheenvironment; and(vii)
compare the industries on the basis of components of TQM.
Thismodel consistsof several components,which representthe attributesand activities
of an organization that is successfully implementing the TQM approach (Fig. 1). The
components aregrouped into four layers:

(i) The top layer consists of the basic features of TQM. These are essentially the
components of the Joiner triangle.
(ii) The second layer consists of the outcomes of the basic features that support
improvement.
(iii)
The third layer is the centre of the model: the ability to control and improve
processes, products and services.
(iv)
Thefourthlayercontainstheoutcomesofthe ability to improve. Thesereˉectthe
twoarmsofthe pincer model: reduced costsandincreased customersatisfaction.
Layers of the S-P model

Layer 1: Fundamental principles of TQM. The fundamentals of TQM principles are thus: (i)
the useofteamprocesses;(ii) the focus on internal customer;(iii) a commonunderstanding
of quality as satisfying the needs of external customer; and (iv) an emphasis on the use of
data and understanding of variationindecision-making.

Layer 2:Prerequisites for improvement. There are three basic requirements that enable an
organizationtoimproveits processes, productsand services.(i)Anunderstandingofcustomer
needs so that the meaning of `improvement’ is clear. Without this understanding, it cannot
be clear whether a change produces improvement or has no et ect on customers. (ii) An
understanding oftheorganization’sprocesses sothattheimpactofchanges canbe predicted


MULTI-ATTRIBUTEDECISIONMODEL 781

and evaluated. This understanding will also require an understanding of the capabilities of
external suppliers and an ability to work with them in improvement activities. (iii) An
understandingoftheprocesses andtechniquesofimprovement so thatpotential improvements
can be identi. ed and implemented et ectively.

Layer 3:Abilityto improve. This is the central focus of the model. If the prerequisites are
present, anorganizationcanimproveits processes and achievethereductionsin waste and
increased satisfaction promised by TQM.

Layer 4: Outcomes of improvement. An organization that can improve can better meet the
needs of its customers. It can reduce the variability of its products and services to provide
greater reliability and can also reduce waste. This in turn reduces costs and makes resources
available for other purposes.

Components of the S-P model

Useof teamprocesses. ``Ateamisagroupofindividuals eachwithspeci.cskillsandknowledge
and interests that enable the members to contribute to the

accomplishment of a common
purpose.’’ The importanceof assuring quality acrossfunctional area was . rst proposedby
Feigenbaum(1956).Hemaintained thatcompaniescould neverproducehigh-qualityproducts
if manufacturing alone had to shoulder the burden of producing quality. Instead,
quality had tobecreated and monitoredbyteamsofpersonnelfromditerentdepartments.

Ateam maybe composedofpeople fromthe samelocal organizationunit orformed
withinagivenfunction.Ateammayhavemembersfromthesame organizationlevel,from
a vertical slice within the organization or function, or a diagonal slice of organization levels
andfunctions. Finally,theteammaybecharacterizedbythescopeofitspurposeand amount
ofits discretiontogather information,makedecisionsandtake action.

Ateamisa processing systemproducing outputs(problemsolutions,decisions,recommendations,
strategic plan, policy statements, engineering drawings, repaired equipment,
etc.) fora userorgroupofusers, e.g.a manager, another team,its ownmembers.Highquality
outputs (innovative solutions) are likely because the idea generating and evaluation
processes used by group (Sink, 1983) can produce results usually not possible simply by
combining the ideas and et orts ofpeople working alone.

Admonishing people to co-operate and help each other will not bringabout the intended
results.Teamworkrequires:(i) a process thatwill overcomebureaucraticcomplexities and
barriers; and (ii) a system of rewards that reinforces use of theprocess.

Several team approaches used for quality improvement are brainstorming, norminal
group, Delphi decision-making, department improvement teams, quality steering teams,
quality improvement project teams, corrective action teams andfacilitators network.

Number of teams registered, number of problems solved to number of problems
undertaken andnetsavinginrupeesby implementingthe improved methodsuggestedby
teams can be considered asindicators to measurethe performance of team activities (Blank,
1995; Capon et al., 1995;Crosby, 1979; Feigenbaum, 1991; Juran& Gryna, 1988).

Focus on internal customers. In a broad sense, a customer can be de. ned as a person who
usesthe service providedby another.Internal customersofa workteam arethe next receivers
of the product or service, who in turn have their customer. All individuals and departments
getinputfromthesource(suppliers); taking theseinputsandprocessing them,adding value


782 R. S. DALU & S. G. DESHMUKH


Figure 2. The customer-supplier relationship.

to the inputs, they receive and provide their customer with a product or service (Fig. 2)
(Gilbert, 1990).

This concept recognizes thateach employee(ordepartment) isa customerfor output
from another employee (or department) and, in turn, has a customer to whom he/she
provides output. With this concept, quality becomes an individual responsibility (Hermann
& Baker, 1985).

If the internal customers’ requirements a

re agreed and met, a chain of quality is made
that reachesouttothe external customers.Togetpeopleto identifytheinternal customers
forthemainoutputsofthe workgroup istomakeastartatatotalqualityprocess.Totouch
ot a dialogue between the internal supplier and internal customer that leads them to agree
customer requirementsisthebeginning of creating a total quality culture.Quality cannotbe
attained externally,untilitisattained internally (Bank, 1995; Zaludova et al., 1998).

Percentage of stat aware ofthe concept of internal customer, percentage of stat satis. ed
withthequality service receivedfromtheirinternal supplierand averagenumberofawareness
programmes conducted in the last3years are considered as indicators for measuring the
componentfocus on internal customer(Feigenbaum, 1991; Juran&Gryna,1988).

Common understanding of quality as satisfying the needs of external customers. To achieve vision
of the company, every employee must know the concept of quality. There should not be any
variation among the employees regarding the concept of quality. In order to make the
employees aware, the company should publish a newsletter on quality and the quality policy
of the company.

Quality awareness can also be improved by involving all employees within the company
in the quality process and getting appropriate input from them.

For manyyears,the practiceoffunctional training was suYcient to enable companies to
make products of a quality which could compete successfully in the market-place. For some
productsthisis stillthecase,butitisnolongerthe caseforsophisticated modernproducts.
So the need is to extend training in quality-related matters to personnel in all functions
(Cheng & Dawson, 1998; Juran&Gryna,1988).

IBM has identi. ed . ve major elements of education for quality, as follows: quality
awareness, quality truths, attitude change through case studies, quality science and methodology
implementation.

Percentage of stat , aware of the concept of quality, ability to understand and express
customers’ needs and averagenumberof training programmes conducted toimprovequality
awareness (last 3 years) are considered as indicators for measuring quality awareness among
employees (Feigenbaum, 1991; Juran, 1989; Swiss, 1992).


MULTI-ATTRIBUTEDECISIONMODEL 783

Emphasis on the use of data and understanding of variation in decision-making. Data means `to
collect suYcient background information to translatetheengineeringproblemstatement into
aspeci.cstatementthatcanbe evaluatedbystatistical method’.

Dataareclassi. edintwoways: `historicaldata’ and `datafromplanned experimentation’.
Historicaldata aredata that are already available. Datafrom planned experimentation are
data gatheredinan attempttostudy theproductionor other problemwhichhas arisenoris
contemplated.

While decision-making,suYcient and accurate information(data)is https://www.wendangku.net/doc/4a16922462.html,eof data
includesplanthecollectio

nof data, collectionofdata, analysethedata, reviewtheconclusions
of the data analysis and present the results.

Data are collected, stored and retrieved either manually or with the help of a computer.
They arecollected regularly or periodicallydepending onthe requirement.For collectingthe
dataandtheiranalysis,varioustoolsareused.Thedataare recordedintheformofrecords,

e.g. rejection record, process-capability records, etc.
Management information programs provide the precise, relevant quality data asguidelines
to managerial and technical actions. The major information areas are: economic data
in the form of quality costs; customer data about product satisfaction; and engineering,
production, inspection and test data about quality levels. Quality information programs
employthedata asafundamentalfactorinthe company andtheplantinformation system,
gearing theinformationtothemeasurementand controlofthe important areasthatimpact
quality control permanent corrective action(Feigenbaum, 1991).

Percentage of employees trained in the use of data (out of employees supposed to be
trained), numberofcasesinwhichthedata areusedfor analysis anddecision-making and
number of times where the past decisions are altered due to the use of data, can be used as
indicatorsfor measuring the emphasis onthe useof data(Cheng & Dawson,1998; Crosby,
1979; Feigenbaum, 1991; Ishikawa,1985;Juran&Gryna,1996,1998).

Understanding customer needs. A customer is someone who is impacted by the product.
Customers may be external or internal. External customers include clients who buy the
product, government regulatory bodies, the public, etc. Within any company there are
numerous situationsin whichdepartmentsandpersonssupplyproductsto each other.The
recipients are often calledinternal customers.

All customers have needs to be met, and the product features should be responsive to
thoseneeds. Thisappliestobothexternal andinternalcustomers. Inthe caseofexternal
customers, the response determines product satisfaction and, in consequence, product
saleability. In the case of an internal customer, the response determines the company’s
competitiveness in productivity, quality, etc. as well as the state of morale among internal
departments.

Discover the needs of customers. Customer needs consist primarily of product satisfaction and
freedomfromproduct dissatisfaction. Theneedsof internalcustomersgo beyond product
andprocesses. Theyincludejob security,self-respect, respectof others, continuityof habit
patterns and still other elements of what is broadly called cultural values. While such needs
are real, they are seldom stated openly. Instead, they are stated in disguised form.

Strategies for determining customer needs. To learn about customer needs, multiple strategies
used arebeing a customer,communicating with customers and simulation(Juran&Gryna,
1988).


784 R. S. DALU & S. G. DESHMUKH

Be a customer: H

ere the supplier carries out some activities similar to those carried out
by customers.

Communicate with customers:Thisisthe most widely used formof discovering customers’
needs.Someofitis initiatedby customerswho haveenduredproductdissatisfaction. Some
isinitiatedby supplierswho areconducting research on productsatisfaction.

Simulation:Simulationistheotherwayofidentifyingtheneedsofcustomersin workshops
and pilot plants. In selected cases this concept ot ers cost-et ective ways of reducing risk.
However, the concept itself is subject to risk as simulation does not fully reproduce the
operating conditions. Hence, it cannot by itself provide appropriate answers to customer
needs.

The ˉ ow diagram, spreadsheet, quality functiondeployment(QFD) and marketing
research arethecommon tools that areused todetermine customer needs.

To measure thesecomponents, three measuresare used:(i)percentage ability toexpress
customerneedsin termsofinternal activities(speci. cation);(ii) percentageofbudgetspent
on identifying customer needs; and (iii) average number of training programmes conducted
toknowcustomerneedsinthepast3 years(Juran&Gryna,1988;Telsang,1997).

Supplier partnership. Therelationshipvariesfrom `adversarial’ to`teamwork’.Inanadversarial
relationshipthesupplieris viewed with suspicion.Intheteamwork relationship,thebuyer
and the supplier work together as if both were part of the same company. This is a planned,
continuing relationship based on mutual con. dence, joint planning, mutual visits and
assistanceD nosecrets.Thesupplieris regarded asan extensionofthebuyer’sfactory.

Tosecurequality improvement,itisvery necessary to achieveateamwork relationship
with suppliers(Miller&Kegaris,1986).To createsuch teamwork,environmenthasrequired
some policy changes,i.e. joint qualityplanning,long-termpurchase agreement, fewer
suppliers, technical assistance to suppliers, published quality policy, suppliers’ seminar,
process controls, incentive programmes, supplier survey and ratings (Feigenbaum, 1991;
Juran & Gryna, 1988).

To understand the status of the relationship with the suppliers, the following measures
canbeused:(i) .nancial and technicalhelp renderedtothesupplierintermsofrupees;(ii)
percentageof suppliers continuedinthelast3 years; and(iii) average numberofseminars
organized (Feigenbaum, 1991; Forker et al., 1997; Juran & Gryna, 1988; Lascelles & Dale,
1988; Mohan, 1996).

Understanding organization processes.

(i)
Documentation :A document is any written or printed information that isgoing to
provide evidence that the company has a system that conforms to the standards.
Theappropriatedocumentationis importantfortwo major reasons:to accomplish
the company’s quality objectives; and to evaluate the quality system by audits and
management review tocreate and maintainacontinuous qualityimprovementloop.
The documentation is the process of preparing these docu

ments to acquire a
standardizedprocessofproduction.Dit erent documents relatingto quality management
may be put into three categories. The . rst category consists of a quality
manual, vision statement and quality policy. The second level ofdocumentation
comprises detailed quality procedures(procedure manual). The third level contains
workinstructions andforms onquality records.
(ii) Problems indocumentation:The problems generally faced are:knowledgeof process;
language; training; . nancial problems; personal capabilities; and management
capabilities.

MULTI-ATTRIBUTEDECISIONMODEL 785

For measuring this component, the measures used are percentage of the processes documented,
percentage of employees aware about the documentation and percentage clarity of
theundocumented featuresofthe processes(Feigenbaum,1991;Juran&Gryna, 1988).

Understanding of techniques of improvement. The techniques for improvement are grouped
into four categories, which are as follows.

(i)
Management methods: The techniques included under this headinghelpthe management
to manage the quality of an organization. Some of the methods are the
Deming wheel (PDCA), departmental purpose analysis (DPA), error proo. ng
(Poka-yoke), cost-bene. t analysis,benchmarking,Pareto analysis,potentialproblem
analysis, QFD, etc.
(ii) Analytical methods: The techniques in this category include cause-and-et ect diagram,
critical path analysis (CPA), fault tree analysis, failure mode and et ect
analysis,Taguchi method, tolerancedesign,etc.
(iii)
Idea generation: The techniques included in this category are used to generate new
ideasin ordertoimprovequality.Theyinclude brainstorming, opportunityanalysis,
idea writing, buzz group, lateral thinking.
(iv)
Data collection, analysis and display:The techniquesunder thisheading areusedto
collect data of a product or a process, analyse them and display them in order to
improve quality. They include bar chart, check sheet, ˉ ow chart, histograms, `P’
chart, scatter diagram, etc.
Number of variousimprovement techniques used, number of training programmes organized
to educate the employeesin use of various techniques of improvement and percentage of
stat trainedcanbeusedas indicatorstounderstand techniquesofimprovement(Crosby,
1979; Mortell& Runger, 1995; Soin,1993;Zaidi, 1995).

Ability to improve. If the prerequisites are present, i.e. quality improvement skills, understandingof
organization’sprocesses, understandingof customerneeds andgood relationwith
suppliers,then an organization can improveitsprocesses and achievethereductionsinwaste
and improve customer satisfaction.

This component can be measured with two measures: (i) percentage understanding of
theknowledgeofhowtechniquescanbeapplied tomajorprocesses; and(ii) averagenumber
of evidencesof improvements(Juran&Gryna, 1988).

Ability to meet the need of the customers. Evidence ofimprovement incustomersa

tisfaction
can represent the ability of the organizationto meet theneed of thecustomer.

Costof warranty(peryear),total numberof complaintsreceivedtosales(inunits) per
yearand percentageriseinsales volumecanbeused as measuresforknowing the ability to
meettheneedsofthecustomers(Juran&Gryna, 1988).

Ability to reducethevariability ofitsproductstoprovidegreaterreliability. Simpli. cation is the
process of reducing the number of types of products within a de. nite range. A little variety
leadsto minimum set-ups and long runs. Simpli. cation enables the Production Department
to improve planning, achievehigher rates of production, machine utilization and simplify
control procedures.

Whenagreat variety exists,asales analysis can be madeto establish the saleability of
theproducts.Whentheaccumulativesalesincomeis plottedagainstthenumberofproducts
ot ered for sale, it is very often revealed that 25% of the product brought in 75% of the


786 R. S. DALU & S. G. DESHMUKH

income. This leads to unnecessary drain of the . rm’s et orts, which can be directed to
promoting the more pro. table and reliable products.

Numberofproduct varietyreduced, percentagerisein salesand machine utilizationdue
toreducedvariability ofproductcanbeusedasindicatorstoknowthe abilityof .rmsto
reduce variability of their products to provide greater reliability.

Ability to reduce waste. Waste refers to any element of production that only adds to cost
without increasing value (Ohno,1991).

Waste can be classi. ed into the following categories.

(A) By nature · Avoidable waste.
· Unavoidable waste.
(B) By resources
Waste of man hours. · Idle time, waiting time, excess of manpower, talent
misuse.
Wastage of materials. · Inadequate systems,ineYcient work methods, low
supply reliability, high defect rates.
Wastage ofM/c hours. · Set-up time, start-up time, M/c breakdown.
Wastage of production · Yielded losses, inspection and testing, over-
facilities and infrastructure. production.
By their causes. · Planning anddesign,purchasing, transportation,

storage, unnecessary stock inventory, usage wastage.
By output. · Reusable waste, scrap, idle cash, expensive
paperwork.

Ability to reduce waste can be measured with the three parameters: number of instances
which show reduction in waste; cost of quality as percentage of sales (COQ); and saving in
rupees per year through reduced waste (Adam, 1994; Chadha, 1999; Feigenbaum, 1991;
Juran & Gryna, 1988; Telsang, 1997).

Development of the model

The AHP has been well received inthe literature and applications of this methodology have
been reportedinnumerous .elds(Mohanty&Deshmukh,1996; Saaty,1980).Thegeneral
approach of theAHPistodecomposetheproblemand tomake pairwisecomparisonsof all
elementsona given level with respect to the relatedelements in theleveljust above. The
schematicofthe modelis showninFig.3.

Description of the AHP

A thorough analysis of the problem is required along with the identi. cation of the important
system attributes involved. Here, the attributes have been selected from the components of
theS-P model,which areasfollows:

Useof team process (UTP)

· Number of team activities (NTA).

· Perceived et ectiveness of team process (PET).

· Outcomes of team activities(OTA).

Focus oninternal customer (FIC)
· Percentage of stat aware of the concept of internal customer (AIC).


MULTI-ATTRIBUTEDECISIONMODEL 787

Figure 3. Schematic of the AHP model.

788 R. S. DALU & S. G. DESHMUKH

· Percentageofstat satis.edwiththequality servicereceivedfromtheirinternal supplier
(SQC).
· Average number of awareness programmes conducted (APC).

Common understanding of quality assatisfying the needsof external customer (CUQ)
· Percentage of stat aware of the concept of quality (ACQ).
· Ability of employees to understand and express customers’ needs (AEC).
· Number of training programmes conducted (TPC).

Emphasisontheuseofdataandunderstanding of variationin decision-making (EUD)
· Extentof training inuseofdata(ETU).
· Extent of application ofdata and analysis (EAD).
· Evidence of impact of variation on decision made (EIV).

Understanding of customer needs (UCN)
· Ability to expresscustomerneedsintermsofinternal activities(AEC).
· Percentageofbudget spent onidentifying customer needs (BSI).
· Average numberof training programmesconducted toknow customerneeds(NTP).

Supplier partnership (SUP)
· Extent of the help(. nancial/technical) rendered to supplier(EHR).
· Percentage of supplier continued (PSC).
· Number ofseminars organized(NSO).

Understanding of the organization’s processes (UOP)
· Documentation of processes (DOP).
· Awareness and understanding ofdocumentation (AUD).
· Knowledge of undocumentedfeatures of processes (KUF).

Understanding of the techniquesofimprovement (UTI)
· Application of various techniques (AVT).
· Training in use of various techniques (TUV).
· Percentage of stat trained (PST).

Ability to improve (ABI)
· Understanding of how techniques can be applied to major processes with the aim of
improvementin customersatisfaction(UTM).
· Evidence of improvement (EVI).

Ability to meet the needs of the customer (AMN)
· Cost of warranty (COW).
· Total number of complaints per year (TNC).
· Percentage rise in sales (PRS).

Ability to reduce the variability of itsproducts to providegreater reliability (ARV)
· Evidenceof reduced variation andincreased reliability(ERV).
· Et ect of reduced variation on volume of sales (EVS).
· Et ect of reduced variation on M/c utilization (EVM).

Ability to reduce waste (ARW)
· Evidence in reduction of waste (ERW).
· Cost of quality (COQ).
· Saving per year through reduced waste(STR).


MULTI-ATTRIBUTEDECISIONMODEL 789

Table 1. Case study

Type of ind

ustry Small and medium scale

Quality status High
Market segment Medium income level
Companygoal Maintain competitive edge

The measuresforeach attributeanditsunitsaregiveninTable6(a-l).

Alternatives

The alternatives are three small and medium enterprises (SMEs) which are to be compared
and evaluated. From the given set of alternatives, the model evaluates the assessment of
status of TQM for the given set of attributes.

Thealternatives are:SSI[O1];MSI[O2]; andMSI[O3].

AHP methodology

The multi-attribute decision model developed here can be used for the assessment of
componentstoTQM.The attributesarecompared with each otherinapairwisecomparison
with respect to the case study described in Table 1. The scale for pairwise comparison and
format ofpairwise comparison matrix areshowninTables7 and8, respectively.The relative
weightsor priorities areobtainedby takingthe opinionsof10experts fromthe . eldofquality
management.The relativeimportanceofeach ofthese attributesisgiveninTable2.From
the analysis,it appearsthattheorganizationO3optionisthebestunderthecircumstances
ofthedeveloped casestudy(Tables3-5).Thereliabilityofthejudgementssuppliedbythe
user canbeestimatedfromFigs4-6which aregenerated betweenthe principalvectorfor
each alternative and its corresponding deciding criteria.

The basic steps in the AHP model

The general structure of the AHP approach that can provide decision supports can be
understood with the help of the following sequential procedure:

Table 2. Weightages for dit erent attributes

Attributes Principal vector
UTP 0.116
FIC 0.068
CUQ 0.140
EUD 0.052
UCN 0.250
SUP 0.010
UOP 0.078
UTI 0.036
ABI 0.027
AMN 0.185
ARV 0.021
ARW 0.017


790 R. S. DALU & S. G. DESHMUKH

Table 3. Weightages of attributes for alternatives

Weight of Weight of

Weight of alternative organizations

sub-criteria criteria
Sr. No. Sub-criteria levelD 3 levelD 2 O1 O2 O3

1 NTA 0.058 0.116 0.167 0.333 0.500
2 PET 0.713 0.116 0.242 0.333 0.425
3 OTA 0.229 0.116 0.264 0.368 0.368
4 AIC 0.251 0.068 0.156 0.375 0.469
5 SQC 0.653 0.068 0.211 0.368 0.421
6 APC 0.096 0.608 0.222 0.297 0.481
7 ACQ 0.260 0.140 0.156 0.375 0.469
8 AEC 0.634 0.140 0.147 0.412 0.441
9 TPC 0.106 0.140 0.200 0.286 0.514
10 ETU 0.100 0.052 0.166 0.500 0.334
11 EAD 0.280 0.052 0.200 0.400 0.400
12 EIV 0.620 0.052 0.215 0.375 0.500
13 AEC 0.686 0.250 0.176 0.353 0.471
14 BSI 0.093 0.250 0.167 0.333 0.500
15 NTP 0.221 0.250 0.215 0.357 0.428
16 EHR 0.686 0.010 0.076 0.386 0.538
17 PSC 0.093 0.010 0.388 0.316 0.296
18 NSO 0.221 0.010 0.250 0.250 0.500
19 DOP 0.665 0.078 0.286 0.357 0.357
20 AUD 0.231 0.078 0.200 0.400 0.400
21 KUF 0.104 0.078 0.130 0.435 0.435
22 AVT 0.571 0.036 0.223 0.333 0.444
23 TUV 0.143 0.036 0.134 0.333 0.533
24 PST 0.286 0.036 0.067 0.400 0.533
25 UTM 0.250 0.027 0.090 0.365 0.545
26 EVI 0.750 0.027 0.186 0.370 0.44

4
27 TNC 0.118 0.185 0.714 0.143 0.143
28 COW 0.380 0.185 0.462 0.308 0.230
29 PRS 0.502 0.185 0.000 0.333 0.666
30 ERV 0.666 0.021 0.250 0.250 0.500
31 EVS 0.223 0.021 0.117 0.295 0.588
32 EVM 0.111 0.021 0.092 0.318 0.590
33 ERW 0.201 0.017 0.227 0.318 0.455
34 STR 0.118 0.017 0.143 0.286 0.571
35 COQ 0.681 0.017 0.462 0.306 0.230

Step 1:Listthe setof organization(Oi,1< i< n).

Step2:ListthecomponentofS-P modelofTQM.Foreach ofthese componentsidentify
thecriteria andthequanti. ableindicatorsforthe criteriaforapossible measure.

Step3: Develop a graphicalrepresentationoftheproblemintermsoftheoverallgoal,the
factors,the criteria andthedecisionalternatives.Such agraphdepictsthe hierarchy
for the problem.

Step 4: Assign weights to each alternative on the basis of its relative importance of its
contribution to each decision criterion. This is carried out through a pairwise
comparison of the alternatives based on the decision criterion. A typical scale for
pairwise comparison which may be used for preparing the pairwise comparison
matrix elements Mkij for each criterion Ck (where Mkij is evaluated when Ai is
compared with Aj)isgiveninTable7.Thegeneralformatofa pairwisecomparison
matrix is also shown in Table 8.


MULTI-ATTRIBUTE DECISION MODEL 791
Table 4. Data summary
Weight of alter native organizations
Sr. No. Sub-criteria O1 O2 O3
1 NTA 0.00112 0.00224 0.00336
2 PET 0.02001 0.02750 0.03515
3 OTA 0.00701 0.00977 0.00977
4 AIC 0.00266 0.00640 0.00800
5 SQC 0.00936 0.01630 0.01860
6 APC 0.00144 0.00193 0.00313
7 ACQ 0.00567 0.01365 0.01700
8 AEC 0.01300 0.03650 0.03910
9 TPC 0.00296 0.00424 0.00762
10 ETU 0.00086 0.00260 0.00173
11 EAD 0.00290 0.00580 0.00580
12 EIV 0.00400 0.01200 0.01610
13 AEC 0.03010 0.06050 0.08070
14 BSI 0.00380 0.00770 0.01160
15 NTP 0.01187 0.01970 0.02360
16 HER 0.00050 0.00260 0.00360
17 PSC 0.00036 0.00029 0.00027
18 NSC 0.00055 0.00055 0.00110
19 DOP 0.01480 0.01850 0.01850
20 AUD 0.00360 0.00720 0.00720
21 KUF 0.00100 0.00350 0.00350
22 AVT 0.00450 0.00680 0.00910
23 TUV 0.00068 0.00171 0.00270
24 PST 0.00068 0.00411 0.00540
25 UTM 0.00060 0.00240 0.00360
26 EVI 0.00370 0.00740 0.00890
27 TNC 0.01550 0.00310 0.00310
28 COW 0.03240 0.02160 0.01610
29 PRS 0.00000 0.03090 0.06190
30 ERV 0.00340 0.00340 0.00690
31 EVS 0.00050 0.00130 0.00270
32 EVM 0.00020 0.00070 0.00130
33 ERW 0.00070 0.00100 0.00150
34 STR 0.00020 0.00050 0.00110
35 COQ 0.00530 0.00350 0.00260

Table 5. Decisionindexforthedesirabilityforeachalternative

Decision Index of O1 0.2055
Decision Index of O2 0.3467
Decision Index of O3 0.4478

Step5: Once the pairwise comparison matrix has been formed for a criterion Ck, normalized
priority of each alternative is synthesized. This is done as follows:

(i)
Sum the values in each column of Mk.
(ii)
Divide each element in the column by its column total, which res

ults in a
normalized pairwise comparison matrix.
(iii)
Computethe averageofthe elementsin each rowofthenormalized comparison
matrix, thus providing anestimateoftherelativeprioritiesofthe n alternatives.

792 R. S. DALU & S. G. DESHMUKH

Table 6a. Use of team processes

Sr. No. Measures
Units

01 Average number of team activities in the last3years (NTA)
Number

02 Perceivedet ectiveness of teamprocess in the last3years (PET) Problems solved(%) of
problems under taken

03 Outcomesof team activitiesinthe last3 years(OTA)
Netsaving(Rs)(%)ofsale
(Rs)

Table 6b. Focus on internal customer

Sr. No. Measures Units
01 Percentage of stat aware of the concept of internal customer (AIC) %
02 Percentage of stat satis. ed with the quality service received from their internal
supplier (SQS) %
03 Average number of awareness programmes conducted in the last 3 years (APC) Numbers

Table 6c. Common understanding ofquality as satisfying theneedsof external customers

Sr. No. Measures Units
01 Percentage of stat aware of the concept of quality (ACQ) %
02 Ability of employees to understand and express customer needs (AEC) %
03 Average number of training programmes conducted in the last 3 years (TPC) Number

Table 6d. Emphasis on theuse of data andunderstandingof variationin decision-
making

Sr. No. Measures Units
01 Extent of training in use of data (ETU) %
02 Extent of application of data and analysis (EAD) Number
03 Evidence of impact of variation on decision made (EIV) Number

Table 6e. Understanding of customer needs

Sr. No. Measures Units
01 Ability to express customer needs in ter ms of inter nal activities (AEC) %
02 Percentage of budget spent on identifying customer needs (BSI) %
03 Average number of training programmes conducted to know customer needs in
the last 3 years (NTP) Number

Table 6f. Supplier partnership/relation

Sr. No. Measures Units
01 Extent of the help (. nancial or technical) rendered to supplier (EHR) %
02 Percentage of suppliers continued in the last 3 years (PSC) %
03 Average number of seminars organized in the last 3 years (NSO) Number


MULTI-ATTRIBUTE DECISION MODEL 793
Sr. No.
01
02
03
Table 6g. Understanding of the organization’ s processes
Measures
Documentation of processes (DOP)
Awareness and understanding of documentation (AUD)
Knowledge of undocument ed features of processes (KUF)
Units
%
%
%

Table 6h. Understanding ofthetechniques of improvement

Sr. No. Measures Units
01 Application of various techniques of improvement (AVT) Number
02 Training in use of various techniques of improvement (TUV) Number
03 Percentage of stat trained (PST) %

Table 6i. Ability to improve

Sr. No. Measures Units
01 Understanding of how techniques can be applied to major processes, with the aim
of improvement in customer satisfaction (UTM) %
02 Evidence of improvement in the last 3 years (EV

I) Number

Table 6j. Ability to meet the needs of the customers

Sr. No. Measures Units
01 Cost of warranty to sale (COW) %
02 Total number of complaints/sales in units per year (TNC) %
03 Percentage rise in sales (PRS) %

Table 6k. Ability toreducevariability ofits productstoprovidegreater reliability

Sr. No. Measures Units
01
02
03
Evidence of reduced variation and increased reliability (ERV)
Et ect of reduced variation on volume of sales (EVS)
Et ect of reduced variation on M/c utilization (EVM)
Number
%
%

Table 6l. Ability to reduce waste

Sr. No. Measures Units
01 Evidence of reduction of waste in last year (ERW) Number
02 Cost of quality as a percentage of sales (COQ) %
03 Saving per year through reduced waste to sale in the last year (STR) %

This resultsinapriority vector PMk 1, which denotes the priorityfor alternative
Ai with respect to criterion Ck.
Step 6: In addition to the pairwise comparisons of the n alternatives, use the same pairwise
comparisonproceduretoset priorities for all the criteriain terms of the importance


794 R. S. DALU & S. G. DESHMUKH


Figure 4. Data summary (O1).


Figure 5. Data summary (O2).

of eachincontributing towardsthe overall goalsoftheorganization.Let Lij denote
each element of the resulting pairwise comparison matrix, when Ci is compared
with Cj.

Step 7: The priority vector PL issynthesized similarly tostep 5(PLi denotes the priorityfor
criteria Ci).

Step 8:Calculatethe overallpriority foralternative Ai denoted by Pi as follows:

Pim 5 dk1 Pmk 3 PLk

5 1

Step 9:Choosethe alternative whichhasthehighestpriority.

Conclusion

The AHP model developed here is useful for assessing the organizations on the basis of
components of TQM. Ithelps toidentify whether the policies and practices of SEMs (or


MULTI-ATTRIBUTEDECISIONMODEL 795


Figure 6. Data summary (O3).
Table 7. Scale for pairwise comparison


Degree of preference De. nition
1 Equally preferred
3 Moderately preferred
5 Strongly preferred
7 Very strongly preferred
9 Extremely preferred
2, 4, 6, 8 Intermediate preferences between the two adjacent judgements
Reciprocal of the above non-If criterion i(Ci) is assigned one of the above non-z ero numbers when it
zero numbers is compared with criterion j(Cj), Cj has the reciprocal value when it is
compared with Cj

Table 8. Formatof pairwisecomparison matrix

Evaluation criteria C1 C2 C3 . . . Cm
C1 1 Reciprocal of entries below the diagonal
C2 Degree of preferences of C2 vs C1 1
C3 C3 vs C1 C3 vs C2 1
:
Cm Cm vs C1 Cm vs C2 . . . 1

large-scale industry) include componentsoftheS-P modelofTQM.It can alsoidentify the
degree to which the various components of TQM are present in the organization.

The index value of each component makes it very clear where the improvement is
needed. So this model not only compares the overall quality performance,

but also guides
for theimprovement programme.

References

Adam, E.E. (1994) Alternative quality improvement practices and organizations performance, Jour nal of
Operations Management, 12, pp. 27-44.


796 R. S. DALU & S. G. DESHMUKH

Blank, J. (1995) TheEssenceofTotalQuality Management (New Delhi, Prentice-Hall of India).
Capon, N., Kaye, M.M. & Wood, M. (1996) Measuring the success of a TQM programme, International


Journal ofQuality and Reliability Management,12, pp.8-22.

Chadha, R. (1999)Quality costs and . nancialperformance:a pilot study, IE Journal, XXVIII, pp. 19-25.

Cheng, P.C.H. &Dawson, S.D. (1998)A studyof statisticalprocesscontrol:practice,problems and training
needs, Total Quality Management, 9, pp.3-20.

Crosby, P.B. (1979) Quality is Free (New York, New American Library).

Feigenbaum, A.V. (1956) Total quality control, Harvard Business Review, 34, pp. 93-101.

Feigenbaum, A.V. (1991) Total Quality Control (New York, McGraw-Hill).

Forker, L.B., Mendez, D. &Hershauer, J.C. (1997)Total quality management in the supply chain: What
is its impact on performance?, InternationalJournal ofProduction Research, 35,pp.1681-1701.

Gilbert, J. (1990) A slice by slice guide to TQM (New Delhi, AYliated East West Press Ltd).

Hermann, J.A. &Baker, E.M. (1985) Teamwork is meeting internal customer needs, Quality Progress, 18,
pp. 12-16.

Ishikawa, K. (1985) What is Total Quality Control? The Japanese Way (Englewood Clit s,NJ, Prentice-Hall).

Joiner, B.L. &Scholtes, P.R. (1985) Total Quality Leadership Versus Management by Control (Madison, WI,
Joier Associates).

Juran, J.M. (1989) Juran on Leadership for Quality (New York, The Free Press).

Juran, J.M. &Gryna, F.M. (1988) Juran’s Quality Control Handbook (NewYork, McGraw-Hill).

Juran, J.M. &Gryna, F.M. (1996) Quality Planning and Analysis (New Delhi,Tata McGraw-Hill).

Lascelles, D.M. &Dale, B.G. (1988) Supplier quality management: attitudes, techniques and systems,
OccasionalPaper 8805,Manchester School of Management,UMIST.

Miller, G.D. &Kegaris, R.J. (1986) An Alcoa-KodakJointTeam, The Juran Report, No.6,pp.29-34.

Mohan, S.R. (1996) Benchmarking:atool toachieve superiorperformance, IE Jour nal, XXV, pp. 20-22.

Mohanty, R.P. &Deshmukh, S.G. (1996)An AHP application for manpower evaluation, IE Jour nal, XXV,
pp. 21-26.

Mortell, R.R. &Runger, G.C. (1995) Statistical process control of multiple streamprocess, Jour nal of
Quality Technology, 27,pp.1-12.

Ohno, T. (1991)ToyatoProduction SystemsD Beyond Large ScaleProduction (Madras, Productivity Press).

Saaty, T.L. (1980)Analytic Hierarchy Process (New York, McGraw-Hill).

Saunders, I.W. &Preston, A.P. (1994) A model and a research agenda for total quality management, Total
Quality Management, 5, pp. 185-201.

Sink, D.S. (1983) Using the nominal group technique et ectively, National Productivity Review, Spring,
pp. 173-184.

Soin

, S.S. (1993) Total Quality Control: Essentials (New Delhi,TataMcGraw-Hill).

Swiss, J.E. (1992) Adopting totalquality management(TQM) to government, Public Administration Review,
52, pp. 356-362.

Telsang, M.T. (1997) Manufacturing excellence through process quality management, 17th All India
Manufacturing Technology Design and Research Conference, REC, Warangal, pp. 205-211.

Zaidi, A. (1995) SPCD Concepts, Methodologies and Tools (New Delhi, Prentice Hall of India).

Zaludova, A.H., Legat, V. &Nenadal, J. (1998)Managing the quality of management training: the role of
the index ofclient satisfaction, Total Quality Management, 9, pp. 5246-5250.