The Control of Quality

8/18/2019 The Control of Quality

1/152


2/152

n

troduction

to the Seri

es

The eig

ht books c

omprising t

he M acmi

llan Handb

ooks in In

dustrial

M anagement series were from the outset planned as an entity, and

togethe

r they cove

r comprehe

nsively ye

t concisely

the varied

aspects

of kno

wledge req

uired by th

ose who m

anage a m

odem fac

tory or

pla

nt. t the

same time

, care has

been take

n to ensur

e tha t each

v

olume sha

ll be comp

lete in itself

, an d carry

sufficient

basic mana

ge

ment th

eory for a

prop er un

derstandin

g

o

f its spec

ific subjec

t.

y

t

his means,

it has been

possible to

avoid a co

mmon pitfa

ll in the

path of

many wr

iters on m

anagement

subjects, n

amely an

attempt

to cover all possible ground in one m ajor volume, with varying

degr

ees of succ

ess.

y

contr

ast, each a

uthor in thi

s series is e

xperienced

in th e sub

ject

o

fhis cont

ribution. A

simi lar pa

ttern has b

een followe

d in each b

o.ok,

bu t eac

h bears th

e stamp of

th e pers

onality of

its author.

Wel l

esta

blished pr

inciples a

nd tested

techniques

are expl

ained, ut

equally ne

w and up-t

o-date con

cepts are e

xplored.

t

is exp

ected tha t

man y prac

tising wor

ks manage

rs and ma

ture

students will wish to have the whole set on their shelves,

ut

tha t

other

s will welc

ome the o

pportunity

of buying

sing le vo

lumes to

m

eet their p

articular n

eeds.

Thank

s are due to

th e auth

ors for the

enthusiasm

with whic

h they

have

joined the

ente rprise

, and to

members of

the staf

f of the

Institut

ion of W o

rks Mana

gers for p

ractical su

pport on

many

occa

sions.

J E

KINS


3/152

M C M IL L N

H N D B O O K S

IN

I N D U S T R I L

M N G E M E N T

Series

Editor:

J

EKINS

ranfield Institute

o

Technology

Executive Editor:

A R PIRIE

Institution

o

Works Managers

FINANCIAL CONTROL IN MANUFACTURING

C. N. Aydon

INDUSTRIAL LAW AND

ITS

APPLICATION IN

THE

FACTORY

Roderick

L Denyer

WORKS ORGANISATION

Alan

Lawlor

EMPLOYEE RELATIONS

WITHIN

THE

FACTORY

W

Mitchell with

Alan

R.

Corbett

THE

NUMERATE

WORKS

MANAGER

H.

R.

Noon

THE CONTROL OF QUALITY

V G. Parry

PLANT LAYOUT AND

MATERIALS

HANDLING

A. W Pemberton

TECHNIQUES FOR PRODUCTION EFFICIENCY

Ken Swann


4/152

The ontrol

o

Quality

V

G P RRY

Macmillan Handbooks

n

Industrial Management

published in association with Macmillan ducation


5/152

©

V

G .

Parr

y 973

R

eprin

t of

t

he or

igina

l edit

ion 19

73

A

ll rig

hts r

eserv

ed. No

part

o

f this

publ

icatio

n m a

y be

rep

rodu

ced

or tran

sm itt

ed in

any

form

or

b

y any

mea

ns

w

ithou

t per

missi

on.

Firs

t publ

ished

973

y

TH

E

M C

MILL

N PR

ESS

LTD

Lo

ndon

and B

asing

stoke

A

ssoc

iated

comp

anies

n N ew

Yor

k Du

blin

M

elb ou

rne J

ohan

nesbu

rg an

d M ad

ras

The

paperb

ack e

dition o

f th i

s book

is so

ld sub

ject

t

o

the

c

onditi

on

th

at

it sh

all no

t , by

way of

t rad

e or o

therwi

se,

be len

t, reso

ld, hi

red ou

t, or

otherw

ise cir

culate

d with

out

the pub lisher s pr io r cons ent, in any form

of

bin ding

or

cover

o ther

than

that in

whic

h

it

is

publis

hed and

w it

hout

a sim

ilar co

nditio

n inclu

ding t

his co

nditio

n bein

g impo

sed

on th

e subs

equen

t purc

haser.

ISBN 978 0 333 14542 5

ISBN 978-1-349-01788-1 eBook)

DOI 10.1007/978 1 349 01788 1

SBN 333 4542 9


6/152

ont

ents

ntr

oduction

to t

he

S

eries

Foreword

y

Richard

Marsh

cknowl

edgements

1 Int

roduction

Aims

and

object

ives

B ackgr

ound

Interchang

eability

D ev

elopm ent

of inspectio

n

D ecline

of in

arrears inspect

ion

Dev elopme

nt of qua l

i ty contro l

Q ua

lity phrase

ology

M

eaning

of certain

terms

Q uali ty

Q uali ty

costs

P

hilosophy

T

he

odds

with which engi

neers co n t

en d

S tatus a

nd relation

ships o

f

qu

ali ty perso

nnel

Presentation

and

format

Q uali ty s

cale

2

Adm

inistration

D esirab

le at t r ibut

es of a qua

li ty m

anager

Typical

industrial

sequence

Re

ports

Q

ual ity

control of

des

ign

Typical

company

organisat ion struc ture

P

ersonnel r

a t ios

Q

ual i ty cos

ts

B u

dgets

1

9

1

2

2

2

2

3

3

5

5

8

8

2

2

22

24

2

6

27

27

28

3

34

35


7/152

Qualit

y audit

36

Packaging

and transi

t audit

39

Fiel

d reports a

nd liaison

40

Travelling and outside reside nt quality personnel 40

Concessio

ns and pro

duction pe

rm its

41

Qua

lity public

ity

4

External quali

ty authorit

ies

43

3 Too

ling and G

auging

Gauge pla

nning and

provisionin

g

44

R

outine gau

ge control

s and syste

ms

45

Metrology laboratory; standards room; jig and too l

insp

ection

47

Tool tr

y-out

5

Press to

ols

51

P

ress tool m

aintenanc

e

54

Jig a

nd tool pe

rsonnel rat

ios

54

4 Raw M

ateria ls

Critical factors

Possibl

e sources o

trouble

Saf

eguards

5

6

R

ange

56

C

ontrols

57

Ferrou

s and non

-ferrous ra

w material

s

57

Paints

9

Pl

astics; synt

hetics; rub

bers

61

Glass

and glass

fibre compo

unds

6

2

Chemic

als and adh

esives

62

D

estructive

and non-d

estructive t

esting

62

S

Goods Re

ceiving Ins

pection

P

urpose

64

C

osts

65

Lo

cation and

space

65

Flow

sy stem

66

Routines and supplier liaison

67

Apprai

sal reports

6

7

Recordin

g o perfor

mance

68

Vendor st

atistics and

qu ality in

dices

7

6


8/152

Sample castings 72

Incoming paperwork 75

Communications and liaison 75

Purchase department interactions

78

Goods inwards inspection procedures: methods and

equipment 79

Sampling procedures 79

Methods

81

Equipment

81

Electronic component problems 83

Electrical breakdown tests 85

Some case

experiences

o

electrical component problems

85

6

Piece

Part

Manufacture

General principles

88

Machine capability 88

First off inspection and inspectors 89

Typical machine problems

91

Lathes

91

Grinding machines

93

Milling machines 97

Gear cutting machines 98

Drills 98

Automatics 99

Numerically controlled machine tools 1

Other machine tool groups

1

Gauging; tooling; instrumentation; instructions 1

Documentation 1 3

Process operations

1 3

Heat treatment

1 3

Painting 1 4

Electroplating; anodising; rustproofing 1 6

Screening: 1 per cent inspection 1 8

7 ssembly

and

Test

Concepts 1 9

Welded assemblies and spot welding 11

Electrode shape

11

Positioning

o

spots 112

Adequate support during welding 112

7


9/152

Clean

lines

s o p

arts t

o be

welde

d

Seq

uentia

l ass

embly

and

in-lin

e test

in g

Fu

nctio

nal a

n d fin

al te

sts

Cleaning

Salva

ge

P

ackag

ing a

n d di

spatc

h

A

ssem

bly p

aperw

ork:

modi

ficati

on co

ntrols

Ele

ctrica

l wir

ing an

d co

nnect

io ns

Sem

i-auto

mate

d asse

mbly

chec

king

Sy

stem

descr

ip tion

P

erson

nel r

atios

8 Re

liabili

ty Te

sting

Desig

n reli

abilit

y and

q ual

ity re

liabili

ty

I

n fant

m or

tality

fa ilur

es

In

fan t m

orta

lity: q

ualit

y mo

nitori

ng tes

ts

Ex

am in

ation

follo

wing

quali

ty cyc

ling t

ests

Test

s by

vendo

rs

Relia

bility

test e

conom

ics

Personnel ratios

9 Ind

ustrie

s othe

r than

Engi

neerin

g

Sco

pe

Paper

m an

ufact

ure

C

uttin

g

K

raft p

apers

Pape

r sack

s

Paperboard cartons

Fu

rnitu

re m a

nufa

cture

Ho

use b

uildin

g

Tex

tiles

R u

bber

Elect

ro nic

s

Chem

icals

Food

s fres

h an

d pro

cessed

Sum m ary

ibli

ograp

hy

In

dex

8

2

2

4

114

1

15

115

1

16

117

119

11

9

122

1

23

1

23

1

24

125

126

127

127

128

128

13

13

1

131

131

1

32

132

133

13

4

134

13

6

1

37

13

7

139

14

1


10/152

or

eword

In the

world

of modern in

dustry it b

ecomes inc

reasingly n

ecessary

for managers to be aware not only

of

the fundamental principles

of

goo

d manage

ment but

also of

th

e latest tec

hniques ne

cessary fo

r

putt

ing those p

rinciples in

to practice

.

W orks

managers

in partic

ular becau

se of the

salient p

osition

which they

hold in t

he manage

ment struc

ture of

m

odern indu

stry

and

their respo

nsibility fo

r translatin

g policy in

to executio

n must be

both

educated

in sound t

heory and

trained in

modern m

ethods.

This s

eries of

ei

ght books

has been d

esigned to

provide th

e basis

of

that education and to supplement essential experience.

I welcome

the opport

unity the I

nstitution

of W orks M

anagers h

as

been g

iven to sp

onsor this

venture an

d commen

d the boo

ks to all

pre sent

and future

managers

in industry

.

RICH

RD

M RSH

Chairman

riti

sh

ail

Preside

nt

Institution

o

Works Ma

nagers


11/152

cknowledgements

The author wishes to thank all who have assisted in the preparation

of

this work and the guidance

of

the publishers.

The management of Rank Xerox Ltd

s

thanked for their permis-

sion and support and also for the use

of

material and details relating

to their quality organisation procedures equipment and facilities.

V G P


12/152

ntrod

uctio

n

AIM S

ND OBJEC

TIV ES

M uc

h h

s

bee

n written

on the sub

ject of qu

ality and

reliability

us

ually cove

ring specia

list proced

ures, stati

stical appl

ications an

d

methods

of

control

primarily s

uited to the

large r orga

nisations.

This b

ook is bas

ed upon p

ractical d

own to ea

rth exper

ience,

rather t

h n any ac

ademic or

theoretical

appr oach.

The purpo

se is to

give

qual ity pr

oduct guida

nce and as

sistance to

the person

nel of both

la

rge and sm

all compa

nies. (The

re are appr

oximately

70,000 com

panies i

n the U.K.

which em

ploy fewer

th n 100

people, and

ove r

8 p

er cent of

the total n

umber of c

ompanies

are in this

category.)

The book may also interest appre ntices and students.

The subj

ect-matter

is presente

d in an eas

ily readabl

e format w

ith

particu

lar empha

sis on the

pra ctical

aspects and

suppor

ted by

examp

les, illust

rations an

d referen

ce materia

l, as ap

plicable.

Inevitably

the conte

nts will be

primarily r

elated to e

ngineering,

but

an eff

ort is made

to

include o

ther indus

tries where

possible.

B

A C K G R O

U N D

o

provid

e an orderl

y introduc

tion to the

subject, i

t is desirabl

e

t

presen

t some his

torical inf

ormation.

t the s ta

r t of this c

entury,

ma

nufacturin

g, and eng

ineering p

articularly

, were still

very m uc

h

h

ighly skille

d craft act

ivities with

qua lity p

rovided by

the skills of

i

ndividuals

, rather th

an a speci

ally organi

sed overal

l system. T

his

arrangeme

nt worked

well und

er the con

ditions of tha

t time

and

high-qua

lity produc

ts were p

roduced, g

iving Brit

ain an env

iable

quality re

putation. (C

urrent qua

lity philoso

phies agai

n recognise

that

quality assurance commences with the people who actually produce

and is e

nsured at th

e time ope

rations are

perfo rmed

.)

Craf

tsmen oft

en produc

ed compl

ete produ

cts, carry

ing out

seque

ntial opera

tions and m

aking part

s

t

o suit eac

h other.

1

1


13/152

INTERCHANGEABILITY

Higher production rates, manufacturing products at separate

locations, and developments such as the motor-car, demanded

interchangeability and spares which would be freely available and

could be used without special fitting or adjustment. Schedules and

tables for limits, fits and tolerances became more necessary and

consequently so did the need for reference length standards.

Interchangeability was made possible primarily through the

brilliant invention by Johansson, the Swedish engineer, of sets of

length blocks or slip gauges and a very ingenious and highly accurate

method for their manufacture. This subject is an engineering saga in

itself about which very little has been written,

or

is known, even today.

Sets of length standards gradually became available

to

all

engineering companies and are now taken very much for granted.

DEVELOPMENT OF INSPECTION

Ever-increasing production rates and high-speed machine tools,

relatively easily controlled, made possible the use

of

semi-skilled

operators, including females. There followed a drift away from

operator responsibility for quality, with continuously more inspec

tion

of

the finished product.

The situation was worsened by the introduction during this period

of

highly sophisticated and accurate measuring instruments, which

were acquired by inspection organisations. Such instrumentation

enabled inspection to become increasingly critical ofproduction, and

to pass judgements from a jealously protected positionofstrength.

DECLINE

OF IN

ARREARS INSPECTION

Progressive managements realised that inspection after the event was

largely negative in principle. It usually involved 1 per cent checking

and the sorting-out of good from bad. Rejects might possibly

e

corrected by costly and difficult rework, but scrap rates could also be

serious. The procedure wasted time, material, equipment and

personnel and also disrupted output owing to the wastage outlined.

DEVELOPMENT OF QUALITY CONTROL

The situation led

to

the introduction of quality control personnel

and concepts. Unfortunately the early quality appointments were

12


14/152

ofte

n quite dis

tinct fro m

the existing

chief insp

ector hiera

rchy, and

this resu

lted

in

frict

ion and slo

w developm

ent.

t

the t ime m

uch emph

asis was g

iven to th

e statistical

approach

and

th is activity became alm ost synonymous with quali ty control.

T

he profess

ional statis

ticians bec

ame deep

ly involved

and

in tr

o-

duced

complicat

ed scheme

s

and

proc

edures, so

m e of w h

ich serve

useful pu

rposes

b

ut

which in

total had

the unfort

unate effe

ct of

con

necting qu

ality con tro

l almost en

tirely w ith

statistics.

G rad

ually, qua

li ty contr

ol functio

ns merged

with ins

pection

organisat

ions

in

a

logical m a

nner , wi th

the deve

lopment of

the

qu

ali ty contr

o l manager

an

d then

the quality

manager,

succeedin

g

the chief inspector as the member

of

the organisation responsib le for

p ro

duc t quality

.

Q

U LITY

PHR SE

OLOGY

Ref

erence can

no w usefu

lly e m ade

to

the

G

lossary

o Terms Use

d in

Qua

li ty ontro

l is

sued by t

he Europe

an Organis

ation for

Quali ty

Contro l.

Th is is a c

omprehens

ive guide o

n most of

the term in

ology

used

in

quali ty contro l

and

is published by E O Q C, Rot terdam ,

Neth

erlands.

Assoc

iated with

the evolut

ionary stag

es outlined

above, a

jargon

of

terms ha

s arisen, in

itially exte

nsively as

sociated w

ith statistic

al

m

ethods

a

nd

latterly on

broade r

concepts.

We have,

for instan

ce,

h

ad single,

double and

sequent

ial samplin

g plans, av

erage qua

lity

level

s AQL) m

erging

into zero defe

cts, quality

assurance

and

total

quality

control . S

ome of th

ese philos

ophies em

brace very

wide

sp

heres of c

ompany o

perations and

are by

no m ean

s universal

ly

accep

ted at this

time withi

n the large

r organisat

ions, to wh

ich they

ar

e principal

ly applicab

le.

A s

stated earli

er, this book

is prim a

rily for p ra

ctical guid

ance and

also fo

r the smal

ler comp an

y,

a

nd

it

i

s not , ther

efore, inten

ded to

dwell ex

tensively on

these phi

losophies.

However, it

is useful

for the

reader to

have som

e knowledg

e of these

concepts and

a very

brief

description

may e he

lpful.

ME

NINGOF C

ERT IN

T

ERMS

A Q L

, or averag

e quality

level, is w

idely used

but

shoul

d

n

ot e

confus

ed with A O

Q

a

nd A O

QL.

3


15/152

AOQ stands for average outgoing quality. t is normally a value

compiled from accepted batches, together with rejected batches,

when the latter have been

100

per cent checked and rejects changed

for satisfactory items. Some variations are practised regarding

replacement ofrejects.

AOQL stands for average outgoing quality limit. (Some people use

the word level instead of limit .) This relates to sampling plan

procedures. Such plans can provide various degrees

of

quality level

according to the characteristicsof the operating curve of the sampling

plan selected. The reader is referred to various books on statistical

quality control and sampling plans for full information on this

subject.

AQL is the level of quality specified by an inspection sampling

plan and normally provides that a small number of defectives can

be

tolerated. (Also, some tests are destructive and cannot

be

applied

100

per cent.) Many people experience concern about any system

which nominally allows a small permitted number of acceptable

defects. They often overlook the fact that

100

per cent inspection is

very costly and can also be shown to have limitations ofperfection, as

compared with inspection

of

a representative smaller sample quantity.

Nevertheless, activities such as space travel demand total elim

ination

of

defects and were primarily responsible for the zero defects

outlook. The principle is that if an AQL

is

allowed, there will

always be some defects. n various walks of life, we do not provide

for, or expect, errors in principle. We expect our salary payments to

be conect and not in error, say, two or three times a year. The same

applies for our bills and day-to-day purchases. Errors do occur,

of

course, but the emphasis of zero defects is that errors should not be

allowed in principle.

The corollary of zero defects is total quality control , which is

still a fairly controversial subject owing to its planned impact over a

much wider field The theory

is

that quality control should not only

apply primarily to manufacturing facets but also to all the other

company functions. Quality representatives should oversee sales,

marketing, engineering, design, purchasing and customer relations

or

service activities. The reader

will

appreciate that such infringe

ment on very jealously guarded domains can generate very hostile

reactions.

n

large companies there is normally very good quality

liaison with all the above departments, but this

is

quite different from

the total quality control concepts outlined.

14


16/152

QU

ALITY

Let us no

w endeavo

ur to am p

lify, qualif

y and defi

ne quality,

and

its i

mplication

s.

The wor

d quality

has a fairl

y clear m ea

ning to pe

ople gener

ally,

whe

ther they

be technica

lly expert

or

layme

n. Further

more, it is

fortuna

te that the

word has

an alm ost

literal tr a

nslation in

to, and

e

qual impor

tance in, m

any other l

anguages.

Un

fortunately

, awarenes

s and reco

gnition of

the need f

or quality

is someti

mes only r

ealised an

d appreciat

ed follow i

ng experien

ce

of

the effect

s

of

poor

quality. M

ost reade

rs will kn

ow

o

f rec

ently

developin

g countries

which acq

uired a rep

utation fo

r shoddy g

oods

but whose go

vernments

now activ

ely and str

ingently en

force goo

d

quality

by ma nda

te, backed

up by lega

l penalties.

As a r esul

t, these

countries

produ cts n

ow carry a

high q uali

ty standing

.

Q

uality is s

atisfactory

confo rm a

nce to spe

cifications

and design

,

suc

h that the

product giv

es cus tome

r satisfacti

on, depend

able servic

e

and

reliab ility

. Optimum

quality

is tha t w

hich prov

ides these

standard

s at th e pr

ice the cust

omer is pre

pared to

pa

y.

QU

ALITY C

OSTS

It

is the

aspec t of

costs wh

ich create

s most m

isunderstan

dings

relative to

produ ct qu

ality.

Firstly, t

he definitio

ns above m

ake it clea

r that from

the c usto

mer

vie

wpoint the

price he

is prepared

to p ay i

s a prime

factor. or

exam p

le,

i

f he req

uires a gol

d watch he

would no

t be satisfie

d with a

stain

less steel o

ne, and w

ould expec

t to pay f

or the high

er quality,

althoug

h both watc

hes may pr

ovide accu

rate time.

Similar e

xamples co

uld be q

uoted for

many oth

er produc

ts:

dome

stic goods,

clothes, fu

rniture, m

otor-cars,

etc. Unsat

isfactory

quality oc

curs when

a customer

pays a pri

ce which sh

ould prov

ide a

corresp

onding sta

ndard,

but receives

an in ferio

r

or

faulty

produ ct

apparen

t initially,

or

which b

reaks down

or

deterio

rates in use

.

The p

roducer m u

st therefor

e ensure th

at he suppl

ies a confo

rming

product an

d provides

adequ ate

quality as

surance co

ver, as app

ro

priate.

Costs of

properly

organ ised

quali ty a

re not jus

t an expen

diture

w

hich is of

doubtfu

l necessity

. Good q

uality cove

r is a so

und

inve

stment and

assurance.

A repu tati

on for poo

r quality ca

n be much

5


17/152

m

ore easily

acquired t

han remed

ied. A cust

omer who

purchases

an

unsatisf

actory pro

duct is un

likely to c

ontinue a

customer o

f the

supp

lier concer

ned. Supe

rior quality

is a decisi

ve factor i

n ensuring

survival over competition when other factors such as price delivery

etc. are s

imilar.

The

se principl

es are usu

ally fully a

ppreciated

by manag

ing and

sal

es director

s. The la tt

er may occ

asionally s

eek low pr

ices to

give

sales ad

vantage bu

t usually w

ith the fir

m proviso that

poor-q

uality

prod

ucts should

n

ot be a c

onsequenc

e.

Buyers m

ay negoti

ate a keen

commerc

ial price b

ut they f

ully

realise that

a low-pr

iced unsa

tisfactory p

roduct wi

ll react to

their

detriment. Somewh at similarly chief engineers an d designers expect

full p ro

duct confo

rmance in

manufactu

re

a

nd tes

t with no

degra

dati

on ofdesig

n standard

s.

It is usu

ally at prod

uction leve

ls

tha

t con

flict ofoutl

oo k arises.

The

production

manager

and his p

ersonnel h

ave target

s to meet

and

quality

personne

l can be an

d often ar

e regarded

as added

expense

or pe

ople who si

mply critic

ise an d hol

d up produc

tion.

Quality co

ntrol are in

a very

invidious

position

at these leve

ls.

They are either the people who hold up production unnecessarily

or w

ho are res

ponsible f

or faulty w

ork havin

g been acc

epted. The

fact that

faulty w

ork origin

ates in p

roduction and

that

quality

contro

l simply fa

iled to ext

ract rejects

is very ra

rely somet

hing for

which prod

uction per

sonnel are

called to ac

count.

Th e

stresses a

re frequen

tly more

difficult w

hen the d

ecisions

in v

olved are s

ubjective o

r as ment

ioned earli

er are due

to the fac

t

that insp

ection usua

lly has bet

te r facilitie

s for fault-

finding. Su

bjec

tive d

ecisions o

n features s

uch as app

earance fi

nish noise

etc. can

be

very m uc

h a matter o

fopinion

. Grou nds f

or friction

on subject

ive

asp

ects can b

e m inimise

d by agree

d physical

samples or

standards

and

w o r d d

escription

s of m eth

od of ass

essment an

d permitte

d

b

lemishes s

uch as scra

tches den

ts etc. Good

quality

liaison is v

ital

to me

et these pro

blems.

Produc

tion peopl

e will quic

kly apprec

iate qualit

y contro l

as an

a

sset rathe

r

t

han a h

indrance w

hen Q

operates th

e three ba

sic

pr

ocedures:

O B S E R V

E

REPOR

T

CTION

16


18/152

This im

plies the

observing of

problem

s, defects,

etc., and

the

reco

rding of

inform atio

n and dat

a to facili

tate the re

porting of

incidenc

e, causes

and possib

le remedie

s, with a

view to

e

ffective

ction

to im prove me thods, facilities, procedures and train ing, as

necessary

. Such proc

edures can

indica te d

ifferences b

etween m a

chine

tools, shif

ts, individ

ual personn

el, etc. Th

e action re

quired may

also

inc

lude reque

st for des

ign to re

view tolera

nces, hea

t treatmen

t,

protective

coatings,

etc.

These

activities a

re the bas

is

of

effec

tive qualit

y control,

which

m

ust always

have the p

rimary obj

ective of e

liminating

the causes of

rejects

so th a t t

hey are u

nlikely

to occur, rat

her than t

he older

inspection procedures

of

ensuring acceptable output quality by

sorting go

od from ba

d. The neg

ative aspec

ts

of

in arr

ears inspe

ction

h

ave alread

y been m

entioned. T

o this can

be ad ded

the f act t

hat

succe

ssive batc

hes of wo

rk can co

ntinue

to have sim i

lar rejects,

unless co

rrective ac

tion has be

en taken.

Once p ro

duction pe

ople realis

e that qual

ity control

can save th

em

pro d

ucing fau

lty work,

avoid scr

ap or re

work, and

high light

personn

el, equip m

ent and m

ethod lim

itations, w

ith suppo

rt for

action, they quickly change their outlook. Their production schedules

im

prove an

d they ten

d to depe

nd more o

n quality

control as

a

pr

oduction a

id. The dif

ficulty at

th

is stage is

that re que

sts can ari

se

fo r un

due extens

ion

o

f qu

ality cover

age to reli

eve superv

ision of

some of h

eir respons

ibilities.

We a

re now abl

e to come b

ack to the

headin g of

his sectio

n, which

is quality

costs. t

should be

readily ap

paren t th

at the cos

t

of

effectiv

e quality c

ontrol of t

he kind ou

tlined can

easily be o

ffset by

the direct s

avings from

redu ction

of

faulty

work, savi

ng

o

f time

and

mate

rial an

d

b

etter ensu

rance of

production

programm

es. These

benefits

can extend

to more

trouble-fre

e assembly

, test resu

lts and

u

ltimately re

liable prod

uct and cu

stomer sati

sfaction.

t

is no t ea

sy to place or

assess th

e savings a

s distinct fr

om expens

e,

whi

ch can aris

e as a resul

t

of

such pr

operly org

anised qual

ity control

.

The

fact th

at

fa

ulty or sc

rap parts a

re minimis

ed can be

shown by

scrap

analysis fig

ures over a

perio d, but

savings

in respect of

overall

im pro

ved effic

iency, mo

re reliable

prod ucti

on delive

ries, and

impro ved field and custom er satisfaction, are difficult to quantify.

These f

actors are

again re f

erred t

o

a

long with

other ass

ociated

featur

es covered

in Cha pter

2

7


19/152

PHILOSOPHY

t may sound the reverse

of

quality thinking

ifwe

pause to consider

that ultimately everything becomes virtually scrap, including

ourselves. The whole scheme of Nature and its cycles has a sequence

of growth, bloom and decline. The time-span

of

these cycles varies

enormously from a fraction of a second for particles

of

matter, to

thousands and millions of years for the universe and geological

features. Similarly some insects live for only a day, and some

animals for over a hundred years.

Nature

is

always busily counteracting the efforts

of

man. A road

built through the jungle will revert with time. Metals corrode and

mechanisms wear out.

These facts should not discourage but should stimulate us in our

efforts to provide optimum-quality products which will defy the

tests

of

time for an appropriate period.

For

example, many motor

cars have been produced, but the outstanding quality

of

Rolls-Royce

cars provides a longer life and continuing value for these cars,

considerably exceeding other makes. (There is, of course, a signi

ficant relative cost differential originally incurred.)

Conversely, some products have

an

initially inbuilt, precisely

defined life-span. Our daily papers suffice for one day only. Similarly,

a printed calender,

or

diary, may cover a period

of

a year, but a

diary which fell to pieces after a few weeks would be of very unsatis

factory quality.

TH ODDS WITH WHICH ENGINEERS CONTEND

Normally there is only one condition whereby a product

is

correct,

or

satisfactory, but there are often, cumulatively, many thousands

of

ways ofpossible deviations, errors or faults in a complete assembly.

Let us consider a very simple bar (shaft, or spindle) which consists

virtually ofonly a length and a diameter. We have stated that there is

only one way in which such a part

is

correct. Conversely, even though

the

part

has only two basic dimensions, there are a number

of

ways

in which errors can be present. A

few

of them are:

Wrong material.

One or both ends out of square to the axis.

Ends out offlat/irregular /dished/domed.

8


20/152

Material hardness incorrect/too hard/too soft.

Diameter tapered/domed/convex (longitudinally).

Bar not straight.

The reader will easily be able to think

of

other possible errors and

can usefully test his thoughts with the additional list tabulated in

the footnote below.

This is one

of

the simplest types

of

piece parts. Immediately one

starts to add further diameters, shoulders, grooves, radii, screw

threads, undercuts, recesses, etc., we meet with additional dimensions

and

possible relationship errors,

and

the odds increase rapidly.

Similarly, acceptable piece parts have

to

be put together correctly,

lubricated, etc.,

and it

will be readily appreciated

that it

becomes

almost impracticable, and would be extremely costly, to inspect for

all these possibilities.

The reader will be very much aware that, fortunately, engineers

have normally already catered for elimination

of

most

of

the possible

variations listed (and thereby reduced the odds) by perfecting

methods and processes, such as true quality control, that reduce the

likelihood of

errors.

For

example, modern precision machine tools

readily produce parallel, smooth surfaces, with square flat ends

and

of

correct dimensions, very reliably and consistently {provided

sensible controls and practices are followed with quality monitoring).

From

the example outlined above, it can be better appreciated why

the concept

of

a permitted number

of

defects,

or

specified QL

(average quality level), is often not readily acceptable. One must of

course stress that the word defect used in this context means a

deviation from specification or tolerance and may not automatically

imply a failure-type defect (e.g. a reject for unsatisfactory appear

ance). In complex electrical assemblies, individu l transistors,

capacitors, diodes, resistors, etc., may have a very low

QL

figure.

When the collective effects

of

a large number

of

very low individual

failure rates are accumulated, the

over ll

failure rate

of

the complete

assembly can be very high

and

completely unacceptable, particularly

as the defect in an electrical piece part can very likely cause mal

function failure. {It is appreciated that the overall failure rate is not

a simple arithmetical addition

of

individual rates. Failures occur

over varying periods

of

time, and mean times between failures

Ovality on diameter; sharp edges on comers; damage marks; finish too

rough/too smooth; diameter too large/too small; length oversize/undersize.

9


21/152

become involved, with failures quoted as hours between failure

rather than percentages.)

STATUS AND RELATIONSHIPS

O

QUALITY

PERSONNEL

Large companies can justify, and require, sophisticated quality

organisations, whereas smaller companies need such cover on

proportionately smaller scales. A very small company of

under ten

employees may cover its quality requirements under the control of

the owner only, and have no staff solely on QC.

As size increases gradually,

we

may have one person responsible

for all quality facets, e.g. porter, signalman and stationmaster . In

such circumstances, the required calibre of the individual can e

equally important to that of a

dep rtment l

QC manager. The smaller

company s QC man cannot justify specialists in every field, or

laboratories, etc., and has to be widely knowledgeable, although

probably in a somewhat specialised field operated by some small

manufacturers.

Whatever the scale, the person responsible for QC needs to have

important personal qualities and abilities, apart from technical

skills. He must always remember that advice is the easiest thing to

offer and the hardest thing to accept.

He must be able to command respect and not resentment.

Prompt and positive decision-making is essential.

A practical down-to-earth approach

is

vital.

Involvement at all levels must be practised.

The quality manager must enjoy suitable status and receive

maximum support from senior management.

In

large companies the reporting responsibility is normally

considered more satisfactory,

if

it is not likely to be influenced by

production considerations. The quality supervisor controlled by the

production manager is not considered desirable but can and does

work well, provided the individuals concerned have the right

personal approach and qualities of

rapport. The more usual arrange

ment

is

line responsibility to a director

or

the managing director. In

some companies the senior quality man has director status, such as

quality director or director ofquality.

The important factor is that quality is an overall concern of every

2


22/152

facet or department and o all personnel within an organisation.

The quality manager is dependent upon every link in a long chain o

responsibilities and understandably the weakest possible links in such

a chain are individual people.

Quality is a continuous activity which cannot be relaxed or

terminated, even when control has been established. All too often,

companies experience problems on product, quality, reject levels,

scrap, etc., and then give maximum support to quality procedures to

resolve the situation. Once problems are resolved, there can be a

reaction, on the impression that future quality has been permanently

assured. Efficient quality departments must always be appraising the

continued need for various quality coverages. This is quite a different

activity from arbitrary cuts and economies imposed by management,

on the assumption that

few

quality problems imply justification for

reduced support o quality activities.

Just as the wise QC manager will introduce new quality activities

gradually and ensure the support

o

all concerned, so should senior

management liaise with the quality manager in ensuring that ease-

ments o quality coverage re based on sound assurance, that

improved methods, controls, etc., have removed the likelihood

o

troubles and will thus allow corresponding adjustment

o

QC

activity.

PRESENTATION AND FORMAT

Many companies perform a service to other sections

o

their industry

by undertaking to provide supplies to customers specifications, as

distinct from direct marketing

o

proprietary products.

n the interest o logical presentation, this book has been written

mainly in the sequence o the average factory which produces com-

plete products, and therefore commences with administration,

tooling and raw materials, and progresses through manufacturing

stages, assembly and test to dispatch. Some chapters on more spec-

ialised subjects follow.

t

is hoped that the various interests will be

covered by some or all

o

these sections. Smaller and direct service

companies personnel will realise that the various remarks and

comments cannot always necessarily apply to the smaller firms

scale and scope o operations, but should find much o interest in the

various details provided.

There is no automatic scheme for quality activities which can be

2


23/152

obtained from charts tables textbooks etc. for application within

particular companies industries or organisations. Each has individual

needs requirements product factors scale market requirements

customer relations etc. which affect the situation.

Furthermore schemes need to be constantly reviewed and amen

ded where necessary to meet changing circumstances and to keep

abreast o technological advance.

QUALITY SCALE

Few companies are able to spell out their quality levels or outlook

in

precise detail. Normally the terms

o

reference for the quality

manager are broadly that he should ensure customer satisfaction

economically rather than costly over-perfection.

Because there is no linear measurement o quality which can be

~

Miraculous

Fantastic

Amazing

l

Outstanding

Exceptional

Excellent

Very good

Good

[

Acceptable

~ f

Fair

Troublesome

e

Poor

[

Extremely poor

Non-existent

Shocking

Terrible

Hopeless

Disastrous

Catastrophic

Fig. 1.1. Quality scale


24/152

readily evaluated or determined, the quality manager is faced with

ensuring that his company operates at the optimum range

of

a very

significant attributes type

of

quality scale, as shown in Fig.

1.1.

Fortunately, the desirable regions

ofthis

type

of

scale occupy a wide

span, as shown in the diagram. The suicidal extremes are more

logarithmic.

No company would logically set out to occupy either of the

extreme positions. The quality manager has a major responsibility

in ensuring that a drift in either direction does not take place.

Managements normally provide adequate support, where the factors

are clear. The quality manager has to provide the information, data,

etc., which will engender maximum support. The level

of

quality is

very much dependent upon the individual quality manager. He has

to

make the guiding decisions, subjective or objective. He has

to

be

seen to be fair, impartial and purposeful.

The situation is very similar to the effects

of

good management

in other walks of life:

1 A bank staff reacts to the example and influence

of

a good bank

manager. The service is prompt, efficient, courteous and con

siderate.

2 Conversely, a restaurant can provide a cold, badly cooked,

limited range

of

food, with doubtful cleanliness and indifferent

service, if the manager is himself indifferent.

n

either instance, the actual relative costs are not directly significant.

The competent manager has

to

ensure that numbers and quality

of

personnel are appropriate and that personnel are trained, instruc

ted and equipped for the duties involved. They must be provided with

satisfactory supplies, materials, etc., and have equitable recompense.

Personnel relations must be satisfactory.

Although custom-built quality assurance schemes for particular

applications are not available as a package, much has been written

on quality practices and procedures which should help those

involved in quality activities and management.

A short bibliography

of

publications on quality subjects is given

at the end

of

this book, for the benefit

of

readers who seek further

information.

23


25/152

2

d

minist

ration

DES

IRABLE

ATTRIBU

TES O

A QUALI

T Y M A N

A G E R

The perso

nal attribu

tes

o

f the

administr

ative head

of

the qu

ality

departmen

t are of p

rime impo

rtance. Hi

s individu

al qualities and

abi

lity to fit in

with oth e

r senior pe

rsonnel ra

te

at

least e

qually wit

h

technic

al compet

ence and

experience

. He need

s to be a

stable,

a

ssured an

d not easi

ly excitabl

e person,

particularl

y when un

der

stre

ss. The ch

oice of ind

ividual to

head a qua

lity depar

tment need

s

very ca

reful cons

ideration a

nd attenti

on, as the

overall su

ccess of

the dep

artment is

very largel

y dependen

t on first-cl

ass leaders

hip.

Unfortu

nately, the

personalit

y

of

an ind

ividual is

almost im

pos

sible to assess

at

an interview. Ideally there should be some depen

dence o

n reliable

personal

recommen

dation, or

proven q

ualities

p

reviously d

isplayed in

a junior p

osition, or

in another

appointm

ent

withi

n the organ

isation. Le

t us list som

e desirable

attributes

:

1 Ability to

communic

ate

at

all le

vels: upwar

ds to senio

r manage

m

ent, o r dow

n to individ

uals, espec

ially withi

n his own s

ections;

also c

rosswise, p

articularly

to

other de

partmenta

l managers

of

equal statu

s, support

ing a bon d

of

team s

pirit with g

ood man

n

ers and g

ood humo

ur. Abilit

y to expre

ss himself

briefly,

clea

rly and w i

th sufficien

t confidenc

e

to

inspir

e respect, b

u t no t

resentmen

t.

2. A

bility

to

c

ompile acc

urate and c

lear repor

ts and info

rmation,

co

nduct cor

respondenc

e associat

ed with t

he appoin

tment,

with

in the p

lant and

with outs

ide suppl

iers, custo

mers,

repre

sentatives,

inspection

authoriti

es, etc. G

ood follow

-up

sys

tem.

3. His personnel should have optimum respect, optimism an d

confidenc

e in him,

apprecia

ting disci

plines imp

osed and

e

xamples s

et, while f

ully able t

o make c

ontact to

report or

discuss dep

artmental

or impo

rtant perso

nal) issues

quickly.

24


26/152

Any executive who becomes too busy to maintain adequate

close liaison with his personnel will lose touch and control.

4

t is sound policy for the head of the department personally

to

interview potential new starters (with the sectional head

concerned, where the overall size of the QC organisation is

appropriate).

Such interviews should include practical tests, as well as

technical questions on Q methods and procedures. The

reader may possibly be surprised by the number of people who

quite blissfully apply for appointments as Q personnel, but

who are unable to use first-principle measuring equipment. In

some instances they have been employed in certain industries

and become expert in a narrow visual check activity and then

assume they have overall quality experience.

5. He must manage his personnel fairly, without bias

or

favour

itism, particularly with respect to salary

or

wage rates, merit

awards, etc. In the larger companies these matters are mainly

governed by

job

description and enrichment, salary and wage

grades, performance targets and appraisals, management

development

and

similar procedures. Nevertheless, the depart

mental head is still able to exercise considerable discretion and

must be seen to be fair and impartial, particularly on these

issues. t is most unwise to make firm promises of future

advancement to personnel, particularly during initial engage

ment.

6

Closely allied to the above factors, most quality managers

become involved in trade union discussions at various levels.

Very special care is essential in such duties. Industrial relations

is a major subject, which is outside the scope of this book, but is

amply covered by various publications and guidance material.

7. The Q manager must keep himself and his personnel up to

date with developments and with new equipment and technol

ogies in his field. Most companies run training or instructional

courses of various kinds and there is no lack of symposia

seminars, courses, etc., on quality control,

both

nationally and

internationally. In fact the main problem is to decide which ones

are worthwhile attending. This kind

of

activity has tended

to

become big business with substantial course fees required.

8. No individual can be a specialist in all fields. The good quality

manager recognises his own limitations and, where required,

25


27/152

enlists the aid of specialists in such fields as electronics, process

controls, chemical and metallurgical laboratories, X-rays and

non-destructive testing.

9 Chairmanship

of

various committees

and

meetings is often

required. Again, this is a subject in itself and outside the scope

of

this book. There must be ability to keep meetings controlled,

brief and purposeful, to issue agenda and write minutes, with

actions, and to ensure follow-up.

10 The head of the quality department needs to be able constantly

and convincingly

to

sell the philosophy, principles, purposes

and mutual benefits ofoptimum quality control.

This list does not claim to include every desirable attribute, but

it

is already sufficiently long for readers to begin to

feel that

birds

with all the above qualities must be extremely rare. Let us therefore

turn to the administrative aspects ofa quality control department.

TYPICAL INDUSTRIAL SEQUENCE

The scale of activities will ofcourse be proportional to the size of the

company. However, virtually all companies operate on varying

degrees of the following sequence:

1

Market research.

2. Engineering design; testing; specifications.

3 Sales; orders.

4. Planning.

5 Purchasing.

6

Incoming materials; supplies.

7. Manufacturing.

8

Processing.

9. Assembly.

10 Test.

11

Packaging; dispatch.

12 Spares; service.

It is appreciated that many other activities may be significant, such

as personnel, accounts, etc.,

but

the main involvement

of

Q

usually occurs in the latter stages, 4-12 inclusive. Smaller companies

on purely subcontract work will be even more specialised around the

production stages.

26


28/152

REPORTS

Finally, under administration should be mentioned the importance

ofa regular report by the quality manager to management.

This activity is additional to normal routine reports and returns.

These special summary reports can e issued at intervals ofnot longer

than a month. Intermediate urgent matters or events should, of

course, be reported separately and immediately.

The quality manager s monthly report summary should preferably

not simply record extensive statistics which are available i required

from normal routines, e.g. the number of batches of work passed

through a department.

The report should be concise, easy to

read and should highlight

features

to

which it is desired

to

draw attention or

to

provide

information.

The exceptions to routine activities are the matters about which

management needs to be informed, e.g.:

1

A very significant increase (or decrease) in the number of

batches of work handled, especially any much higher defect

or

rework experience.

2. Serious quality problems with particular suppliers.

3. Adverse field experience.

4. High scrap or reject experience.

5 Personnel problems, etc., wages, salaries, trade union pressures.

6

Serious delays in any areas.

7. Inter-departmental difficulties.

Management will appreciate such communications from the

quality manager and will usually liaise, support and follow up, as

necessary, on the information provided.

QUALITY CONTROL

O

DESIGN

Much has been written about quality control of design, recom

mendations on practicability of tolerances, machine capability

studies and other specialised techniques. While these more sophis

ticated procedures are practised (mainly in the larger companies),

they are not generally widespread. Further comment will be made on

them in later chapters.

As a general rule, the QC department s involvement begins with

the planning stage. Gauging, facilities, equipment and inspection

27


29/152

methods require consideration and action. Large companies enjoy

gauge planning and inspection methods personnel. n a small

organisation the quality manager plans these activities.

Next comes liaison with purchase and supplier appraisal, followed

by receipt of raw materials and bought-out finished supplies. These

features will be covered more fully in Chapter

5.

There follows the manufacturing, assembly, test and packaging

stages, all of which will also be covered in later chapters. The last

stages of spares and service will normally directly affect only those

companies marketing their own proprietary products. Nevertheless,

these are again important areas which will be covered later in this

chapter under the heading

of

field

reports and liaison (seep. 40).

TYPICAL COMPANY ORGANISATION STRUCTURE

Fig. 2.1 illustrates very briefly the relationship ofa quality department

in a company and the basic departments in an average quality control

organisation.

Some companies (mostly small in

size

still do not operate a

recognisable quality activity.

Depending

on

a company s size and type of product, this basic

structure can be amended as appropriate. t will normally show

titles such as quality manager, his deputies or assistants and the

respective departmental supervisors, who may be variously called

foremen, assistant managers or quality engineers. Except in the

very small company, the quality manager s importance should merit

his having an efficient personal secretary. Such services provide

immense help with administration activities and general organisation.

n

setting up his organisation, the quality manager must keep

uppermost in his mind that the personnel outlined above are

primarily administrative and rarely carry out actual inspection or

quality control duties. Too much supervisory cover is very unlikely

to

ensure good quality production. The production or assembly

floor involvement of quality rank-and-file personnel is the essential

requirement of efficient quality control. These are the areas where

the supp rt

ofquality people has the main company impact and value.

n

the smaller companies where the total people engaged on quality

control are few in number, such problems should not arise.

Where the quality manager has a department of such a size that it

merits deputies and foremen, he must ensure that such personnel

28


30/152

S

N

I

O

R

C

M

P

E

X

E

U

T

V

E

m

a

n

e

c

o

r

s

e

d

e

o

g

a

m

a

o

r

s

m

a

e

c

m

p

p

o

e

o

A

d

m

i

n

Q

u

y

A

u

F

e

d

a

V

e

L

a

s

o

O

u

s

d

p

o

S

e

a

y

F

g

2

1

T

q

y

d

p

m

e

n

n

a

c

o

m

p

n

n

d

s

s

e

o


31/152

h

ave

th e

ap

p ro

pria

te t

ech

nica

l c

om p

ete

nce

a

nd pe

rso

nal

q ua

liti

es,

a

nd co

nsi

dera

ble

c ar

e m

u st

e tak

en

w he

n a

n y

such

ap

poi

n tm

en t

s

a

re m

ad

e.

Similarly, the relative s ta tus

of

such personnel m u st be clearly

def

ined

fo r

a s

atis

fac

tory

a

nd i

mpo

rta

nt de

par

tm e

n ta

l te

am

spir

it to

p

rev

ail.

P

ER

SO

NN

EL

R

A T

IO S

Th

e abo

ve

com

m e

nts

tou

ch

indi

rec

tly

upo

n th

e su

bje

ct

of ra

tio

s o

f

nu

m b

ers

of

per

son

nel.

I ns

pec

tion

o

r qu

alit

y c

on tr

o l i

s

an a

ctiv

ity

w

here

it

is a

ll

t

oo eas

y to

en

gag

e

to

o m an

y p

ers

onn

el.

A ll

su

ch

peo

ple

w il

l a

ppe

ar to

be

ac

tive

ly i

nsp

ecti

ng

an

d m o

st

likely o ver

ins

pec

ting

b y,

e .g

., r e

peti

tive

o

r m u

ltipl

e in

spe

ctio

n o

f the

sa

m e

wo

rk

in

di

ffer

ing

loca

tion

s.

Thi

s is

a m

o s

t un

des

irab

le s

itua

tio

n w

hich

c

an gi

ve r

ise

to s

eve

re

cri

tici

sm .

To

o

m

u c

h

insp

ect

ion

res

ults

i

n crit

ical

s ta

nd

ards

and

hig

h r

ejec

t le

vels

, w i

th c

orr

esp

ond

ing

and

ju

stifi

able

re

sent

m e

nt

b

y

a

ll c

onc

erne

d, p

art

icul

arly

p ro

d u

c tio

n pe

rso

nne

l.

Sim

ila

rly,

i ns

pec

tion

act

ivit

ies

sho

u ld

be c

ont

inu

ous

ly m

on

itor

ed ,

otherwise s om e work will continue long af ter the n eed has d isap

p

eare

d .

Th

e sa

m e

com

m e

nts

also

ap

p ly

to

p ap

erw

ork

sy

stem

s

an

d

rec

ords

.

U

nfo

rtu

nate

ly t

her

e is

no

sim

ple

or

r

elia

ble

w ay

o

f ass

essi

ng i

dea

l

ra t

ios

o

f nu

m be

rs

of

qua

lity

pe

rson

nel

. W

ith

ou

t sig

nif i

can

ce

of

im p

orta

nce

in the

re

lativ

e o

rde

r li

sted

, so

m e

o

f the

m a

n y

fac

tors

in

vol

ved

ar

e:

Siz

e o

f co

m pa

ny.

T

ype

of

p

rod

uct

.

V a

riet

y o

fpro

duc

t.

Q ua

ntit

ies.

M a

nufa

c tu

rin

g m e

tho

ds.

C on

d iti

on

of eq

u ip

m e

nt (m

ac

hine

s, p

roc

esse

s).

T

oo l

ing

an

d gau

gin

g.

C

las

s o

f l

abo

ur:

skil

led/

uns

kille

d/m

ale

/fem

ale

;

labo

ur t

u rn

ov e

r/

t

ra in

ing

.

Sh

ift-

w or

k pr

obl

em s

.

P

rop

ort

ion

of n

-ho

use

m a

nufa

ctu

re .

C o

m p

an y

qu

alit

y sta

nd

ards

; at

t i tu

d e ;

s up

por

t.

D e

sig

n st

and

ards

; sp

eci

fica

tion

s; c

om p

lex

ity;

to le

ran

ces.

30


32/152

In some companies raw material supplies are controlled

by

chemical or metallurgical laboratories which are headed

by

the chief

engineer. Similarly final test personnel are also sometimes con

trolled by the chief engineer. In other companies these people are

part

o

the quality department. These are just two instances apart

from the above list o factors

why

ratios

will

vary markedly between

different plants and companies.

Additionally the method of computing ratios may vary widely.

Inspection or quality personnel are most commonly expressed as a

ratio o actual numbers on production i.e. people actually making

the products direct operating personnel. To be realistic this ratio

should exclude the administrative staff in the QC department and

count only those actually carrying out physical inspection duties.

However it

is

normal to include the whole o the QC personnel

when assessing ratios; hence the importance of keeping QC admin

istrative personnel closely controlled in numbers as stressed

previously.

The reader will no doubt still expect some guidance on ratios but

this guidance can only be

very

approximate for the reasons outlined

above.

Ratios can vary from one inspector to four producers in an

industry such as aero-engines and conversely to one inspector to

forty producers in industries where the nature

o

the product

is

less

critical or the processes are automated and very unlikely to require

other than infrequent sample monitoring.

These would normally be considered the extreme ranges. Always

the aim should be to place the responsibility for quality with the

people actually producing the product. Quality personnel primarily

monitor the product and should not relieve the producers from their

responsibility for ensuring a satisfactory product.

The reader will obtain some further guidance

by

noting the

concentration

o

ratios which roughly prevails to one side

o

the

mean o the range as shown in Fig. 2.2.

Normally a general machine shop

is

an area where the need for

most inspection

o

quality control

is

necessary. Parts are being

originated from raw material and faults can easily mean costly

scrap or rework and delay. Conversely in assembly areas the

final

function and performance tests can be sufficient to monitor product

quality adequately. Failure o product to pass final test can

usually be resolved by further adjustment or assembly correction

3


33/152

b

acked

up y

a

ction

to p

reven

t rec

urren

ce a t

the

sourc

e

o

f

the

tro

uble.

It

is app

recia

ted th

at so

me r

eader

s will

know

o

f a

reas

where

the

above general statements are no t applicable.

or

example, some

com

panie

s use

high

ly sk

illed

mach

ine-s

hop o

perat

ors u

pon

whom

they

are

able

to d

epend

for

qual

ity, w

ithou

t ext

ensiv

e cov

er by

qu

ality

contr

ol.

Som

e pro

duct

s,

if

w

rongl

y ass

emble

d, ca

n be e

xtens

ively

dama

ged

M

ean o

f ratios

1:5

Fig 2

2 Ra

tios of

quali

ty con

trol pe

rsonne

l to n

umber

o

f pro

ducer

s

du

ring

final

tests.

Henc

e the

need

, in

such

instan

ces,

for st

ringe

nt

sub-

assem

bly q

uality

cont

rols.

C

learly

, in

any i

ndust

ry a

great

deal

depe

nds u

pon

the q

uality

ma

nager

, his

senio

r per

sonne

l or

both

, in e

xerci

sing

carefu

l and

con

tinuo

us m

onito

ring of

in

spect

ion d

uties.

Skil

l and

acum

en i

n

this

field

, clos

ely a

pplied

in

the d

ay-to

-day

activi

ties,

i

s

the

only

posit

ive w

ay of

ensu

ring s

atisfa

ctory

ratio

s. Th

e QC

man

ager

must

no t

e

an empire-builder .

Th

ere ar

e dan

gers

in app

lying

too

sophi

sticat

ed me

thod

s, suc

h as

in

crem

ental

times

for

each

step i

nvolv

ed in

inspe

ction

, espe

cially

if

m

ultipl

icatio

n is c

arried

ou t o

n ini

tial ap

prox

imati

ons:

3

2


34/152

10

minutes to obtain drawing and paperwork

90 minutes to inspect

10 minutes to write report

10

minutes for discussion

on

results

120 minutes 2 hours

If one then proceeds to multiply such data by, say,

500

batches of

work in a week, one obtains a total of 1,000 hours, equivalent to

25 personnel

on

a forty-hour week. Even this has

to

be amended fot

an

average (say)

80

per cent effective time utilisation, resulting in 31

inspectors apparently being necessary. This is usually rounded off

to 32 and could be repeated for several departments. The result

would be gross overstaffing. The reader will readily

see

the

flaws

in

such a system.

Another fallacy often put forward is that

if

the rate

of

production is

varied, the inspection

or

quality personnel should vary directly

proportionately. The quality manager who adopts this argument

must be prepared for it

to

operate both ways.

If

it is contended

that

because output is doubled, twice the number

of

inspection personnel

are required, then conversely a halving of output should justify a

50 per cent cut.

Obviously, increased output results in better production tooling

for larger batches and longer runs, and could mean only marginally

more inspection. Alternatively, more machine tools could be

involved, or additional shift working. The effect of these latter

measures would have a more significant impact on the amount

of

quality coverage required.

These factors serve to illustrate the difficulties in attempting to

specify inspection ratios by arbitrary means, and emphasise the

importance of the integrity and expertise required from the quality

manager in establishing his personnel requirements relative to his

particular company.

In the context of goods inwards and tooling inspection, there are

further special factors

in

connection with personnel ratios which will

be dealt with in Chapters 3 and

5.

Finally, the continuing advances in semi-automated processes and

equipment have a bearing on this subject, and some examples of

new developments

on

these lines are dealt with in Chapters 5 and 7.

T.C.O.Q. B

33


35/152

QUALITY COSTS

Staffing ratios are directly connected with quality costs. In this

respect, simple personnel unit numbers can sometimes be varied. For

example, it may be practicable to utilise higher numbers of semi-

skilled personnel and juniors on certain work rather than small

numbers

ofcostly, highly skilled personnel.

Much has been written on methods

of

assessing the value

of

quality, ranging from the costs

of

actual appraisal

or

monitoring

through fault prevention, corrective action, costs of scrap, service

and warranty claims. Transcending all these in importance is the

ultimate target

of

optimum customer satisfaction.

Unfortunately, unless a company has actually and painfully

experienced the effects of poor overall quality, it is usually difficult

to allot meaningful criteria and values in these areas. A

few

examples

follow.

A reduction of a very high scrap rate by additional expenditure on

QC can look most impressive and could no doubt be fully justified.

However, the improvement could also be due to better facilities,

prompted by QC effort. Indefinite continuance

of

the same higher

level

of

incidence

of

QC cover may not be necessary.

As

stated in Chapter

I

an optimum acceptable level

of

quality is

High

f igh

Fig.

2 3

The acceptable balance point between quality costs and incidence of

failures

34


36/152

the c

orrect targ

et . The co

sts of qual

ity acceler

ate rapidly

according

to

the ne

arness of a

goal ofabs

olute perfe

ction.

Fig. 2.3 il

lustrates that

there c

an be a p

oint where an

acceptab

le

balance occurs between quality costs

and

the incidence

of

quality

failures re

jects, sc rap

, returns, et

c.).

t must be e

mphasised

that

bet te

r quali ty can

often b

e achieved

w

it out a

correspond

ing increas

e

in

costs.

Rejects ca

n be reduce

d by

b

etter staff

instructio

n, organis

ation,

or

methods and

facilit

ies

witho

ut significa

nt cost pen

alties. Ob

viation of p

iece part f

aults can

minimise

assembly

problems and

ensure

satisfacto

ry functio

n

o

n

final test.

UD

GETS

Pers

onnel rati

os

a

nd qu

ality costs

lead logi

cally into

budgetary

contr

ols. Virtua

lly all modem

com p

anies ope

rate on a

budgeted

costs ba

sis. O ne of

the more

impor t an t d

uties of th

e qu ali ty m

anager

is to pr

epare his d

epartmenta

l budget, s

ubmit

it

and

obta in ap

proval

f

rom mana

gement. U

sually bud

gets are on a

n annu

al basis, w

ith

periodic reviews

and

fu ture forecasts.

To

enable im to

prepa

re his budg

et, the qu

ali ty mana

ger mus t

have de

tailed advic

e from m a

nagement o

f product

ion progra

mmes,

a

ctivities and

anticipate

d new deve

lopments .

T h

e pr imary

sections of

a budge

t are norm

ally perso

nnel, test

exp

enditure,

capital pla

n t

a

nd eq

uipment, f

urniture and

fittings

,

travel and

trainin

g . Then p

ossibly m o

re inciden

tal items s

uch as

s

tationery ,

printing,

etc., which

m a y

or

may not b

e covered

by

in

dividual de

partments

in some co

mpanies.

In m o

st instances

personnel

costs form

the bulk of

the expen

diture

of

a quality

departmen

t. Hence

the need fo

r the caref

ul control o

f

numb

ers emphas

ised earlie

r. In asses

sing his p

ersonnel c

osts, the

quality m a

nager mus

t allow for

wage and

salary adv

ances, over

t ime

and

night-s

hift prem i

ums. High

incidence

o

f overtim

e

and

nig

ht

s

hift wo rkin

g

can

add

substantial

ly

to

the to

tal . Hence

the need f

o r

inf

ormat ion on

producti

on program

me plans.

A n

ticipated

increases

or decreas

es) in per

sonnel str

ength are

us

ually requ

ired

to

ind

icate their phasing-in over the period being

budg

eted. Thi

s is logica

l, as an e

ffective bu

dgetary s

ystem will

prov i

de for mon

thly return

s from acco

unts to dep

artmental

managers

reporting

progressiv

e budgetar

y experienc

e.

5


37/152

In some co

mpanies t

est expend

iture can b

e the next

largest figu

re

to

pe

rsonnel. T

his activity

is also c

losely rela

ted

to

pr

oduction

programm

es. Howe

ver, the q

uality ma

nager need

s carefull

y to

review test procedures periodically and especially

at

budget times.

Test e

xperie

Documents

The Control of Quality