dr-fb-rs-cs rrpmom 40 (2011) pre-print

8/2/2019 Dr-fb-rs-cs Rrpmom 40 (2011) Pre-print

http://slidepdf.com/reader/full/dr-fb-rs-cs-rrpmom-40-2011-pre-print 1/9

1

An Application of the SMED Methodology in an Electric Power Controls CompanyDomingos Ribeiroa, Fernando Bragaa, Rui Sousab, S. Carmo-Silvab

a General Electric Power Controls Portugalb Production and Systems Department, Engineering School, University of Minho

Postal address: Dr. S. Carmo-Silva, Production and Systems Department, Engineering School,

University of Minho. Campus de Gualtar, 4710-057 Guimaraes, Portugal.Email: [email protected]

AbstractLean production is a strategy for high competiveness in manufacturing. The capability for economical

manufacture in small batch sizes is an essential requirement for achieving lean production. This facilitates mixed

production of several kinds of products to match varying product demand and can have a major impact in

reducing inventories. An obvious requirement for this is the high frequency of equipment setups or product

changeovers. This will not be attractive unless set-up times and costs can be reduced to competitive levels. The

application of SMED can achieve this. SMED is a well-established methodology involving a set of techniques,

methods and guidelines to achieve fast product changeovers at machines. This paper describes the application of

SMED in the production process of plastic and metal components required for the assembly of several types of

circuit breakers. The work was carried out during a short period of a few months under a master thesis project.Several important SMED strategies and solutions were implemented and evaluated in terms of their impact on

productivity and on other manufacturing performance measures. Three specific machines were involved: a

punch-bending machine, a punch press and an injection moulding machine. An important contribution was made

by introducing innovative and simple solutions such as adapting tools and normalizing changeover operations.

Most of the achieved results exceeded the initial expectations. Beyond the purely technical and economic

benefits of SMED, better workstations’ ergonomic conditions were also attained. Besides the usual

quantification of setup time reduction, other indicators were calculated, namely: work-in-process (WIP), annual

setup cost and distance travelled by operators during the changeover process. Reductions of setup time varying

from 59% to 90% were achieved. WIP of metal components was reduced from 17.05 to 7.74 days reducing more

than 50% on the corresponding costs. A more impressive reduction on WIP was obtained for plastic parts,

actually from 5 to 1.09 days of work corresponding to a WIP cost reduction of over 80%. The distance travelledby operators during the changeover process was dramatically reduced too: typically a reduction from 300 m to

10 m and less. The total annual cost savings projection, in this small area of parts production, is near 20,000 €.Although large benefits were obtained from the study, scope for further improvement still exists. In fact the

objective of product changeover times below 10 minutes aimed by SMED was not achieved in one case.

KeywordsMachine setup, Quick changeover, SMED, Mass customization, Lean manufacturing.

IntroductionIt becomes increasingly evident and is generally accepted that mass production, which has dominated the

industry for a long period, in the last century, has become obsolete [1]. More than ever, companies must be able

to manufacture a large variety of products, in small quantities, in order to respond to market requirements [2].

However production costs must be competitive, otherwise a company may not be able to subsist in the current

economic markets within the global competition paradigm of today. Well known new manufacturing paradigms,

namely lean and agile manufacturing, emerged to cope with the challenges of competition within the new marketparadigm. Among other aspects, it is obvious that the necessary time to change production from one product to

another, commonly referred as product changeover time, must be kept as short as possible for allowingmanufacturing in very small quantities of a great variety of products at competitive production costs. This is the

purpose of the SMED methodology, one important tool initially though for lean manufacturing but of great value

to agile manufacturing. SMED aims to substantially reducing product changeover time towards achieving single

digit time values, i.e., less than 10 minutes.

The main objective of this paper is to describe an industrial application and implementation of the SMED

methodology in a company of electric power controls, more specifically within a production process of plastic

and metal components involving three machines: a punch-bending machine, a punch press and an injection

moulding machine. Besides the usual measurement of the setup time reduction, a clear objective of this work isthe quantification of other indicators, such as: work-in-process (WIP), annual setup cost and distance travelled

by operators during the changeover process. The annual setup cost is of particular importance for the company

administration.



Tfundame

the electr

fourth se

Fundam

Ldevelope

specially

chain of

identifies

inventor

waiting.

wasting

comprom

W

manufact

more an

requires

changeosuccess [

S

changeo

changeogood par

equipme

intangibl

identify

Intangibl

and equi

A

knowled

and, (iii)operators

equipme

cannot b

processe

therefore

serious o

S

if the equ

methodol

internal

is consid

many ch

when theoperationlast stage

Standardi

1 demon

die heigh

e structuretals on lean

ic power con

tion. On the

entals of Lea

an manufactd at Toyota

focused on t

a product. W

seven types

, transport (

The expressi

less time, u

ising the pro

ith today’s

uring philos

more impor

supplying s

ers. So, freq11, 12].

ED is a m

ers of produ

er is a produin a given b

t and tools,

. The first c

he following

e costs inclu

ment stress.

cording to S

e about the t

knowledge’ motivation

t designed f

e achieved if

[14]. It is al

unable to id

bstacle to SM

ED classifie

ipment is sto

logy compreh

perations int

red the most

angeover pro

equipment is are thorougis the rationa

ization of dim

trate the use

t (320mm).

f the paperroduction an

rols compan

last section s

n Manufact

ring [3, 4] isotor Compa

e identificati

aste is consid

of waste:

of materials)

on “less is

sing less pr

ucts’ quality

ressure for

phy, improv

tant [10]. As

all quantitie

ent product

ethodology

cts. The met

ction set-upatch and the

setups, runni

an be expres

tangible cos

e the costs i

kine and Ar

chnical aspe

about the mfor collabora

r quick setu

the operator

so recognize

ntify profits

ED studies a

s the change

ped, and ext

ends three s

external op

important [1

esses observ

stopped thahly analysedlization of all

ensional asp

of shims atta

is as followd SMED. Th

. The main r

me concludin

ring and SM

the western dny after the

on and elimi

ered any acti

verproductio

, movement

ore” represe

duction are

cost and deli

reducing no

ing performa

previously r

s of a great

hanges and s

ut forward

hodology is

activity extenirst good par

g of trial pa

ed in monet

ts: lost produ

herent to: re

i [11], the ef

ts of equipm

thod (how tting and acc

s and an ad

s do not coo

that many o

associated to

d implement

ver operation

ernal operatio

ages: (i) sep

rations (whe

], and can le

ed in the ind

could be exowards convthe operation

cts is one of

hed to dies to

2

s. After thise third sectio

sults of the i

g remarks ar

ED

esignation foecond Worl

nation of was

vity or event

n (in terms

(of people),

nts the idea

, as well a

very [9].

n-adding val

nce acting u

ferred, the t

variety of

hort product

by Shingo [

upported by

ded over thet of the follo

rts and adjus

ry units whi

ction, lost c

sponse time t

ficiency of ch

nt and tools,

o do). An ipting new w

equate work

perate and r

rganizations

the changeov

tions.

s in two type

ns, which ca

aration of in

ever possibl

ad to a chang

ustry, it is c

cuted with trting them is (both inter

he issues fre

obtain both

introduction,n describes th

mplementatio

outlined.

r TPS - ToyotWar [8]. It

te, also know

that does no

of earliness

over-processi

of the Japan

s less mater

ue activities,

on products'

pe of deman

products, lik

changeovers

] focused o

a set of tec

time intervaing batch, a

ments [13].

le the second

pacity, inven

o client, risk

angeover is d

(ii) work org

portant reqorking practi

organization,

cognize inhe

o not take i

er time reduc

s: internal op

be executed

ernal and ex

), and, (iii) r

eover time re

mmon to id

e machine rto external oal and extern

uently addre

standard cla

the seconde industrial a

n are present

a Productionis considered

n as “muda“,

add value t

nd/or exces

ng (or inad

se productio

ial and hum

which mea

changeover

d that charac

ly demandi

become key r

achieving

niques, meth

l between thed, typically,

hangeover c

cannot [16].

tory, labour,

of market sh

etermined by

nization (wh

irement fores and meth

efficient and

rent advanta

nto account c

tion [15]. Thi

rations, whi

while the eq

ernal operati

tionalization

uction of 30

ntify operati

nning [20]. Ies whenever

al) to make th

sed in chang

mping height

section provipplication of

d and discus

System [5, 6,a systematic

, along the v

the product.

sive quantity

quate proces

n philosophy

an resources

ns to imple

activities is

terizes curre

g a great n

equirements

very quick

ods and gui

productioninvolves pre

osts can be t

Yang and

trial parts an

are loss, and,

three key el

o does what a

the SMEDods. In fact,

effective ch

es of new c

hangeover c

is fact may c

h can only b

ipment is ru

ons, (ii) con

of operations

to 50% [19].

ons that are

n stage (ii) tpossible. Them more effi

eover process

(80mm) and

des someSMED in

sed on the

7] mostlyapproach

lue added

Ohno [5]

, defects,

sing) and

, meaning

, without

ent lean

becoming

t markets

umber of

o achieve

roduction

elines. A

f the lastaration of

angible or

eane [17]

d defects.

operators

ments: (i)

nd when),

uccess iseven with

angeovers

angeover

sts, being

nstitute a

executed

ning. The

ersion of

. Stage (i)

In fact, in

erformed

e internalthird andient [18].

es. Figure

a standard



T

and time

that allo

of centri

T

studies th

of functi

be point

operation

that can

these and

SMED cT

results an

In

reportedperform

internalstorage.

quite far

reduction

B

here rep

in a redu

W

companicapacity

demand

dependenchangeo

frequent

finish an

of invent

handling

variation

the mach

importan

compani

and oppo

SMED a

Tcompany

e correct pla

consuming t

quick setup

g jigs applie

e cases repre

at can drama

nal clamps, i

d up the i

s of the sam

bring about

other SMED

se studiese SMED me

d reports of t

the area of

by Kays anddetailed ana

and externalhe main ach

from the sin

s in product l

ian [23] desc

rted. A chan

tion of the m

ith a succes

s can be achiand the prod

[24]. It is re

t on changeoers to avoid

product chan

WIP levels

ory turnover,

and increase

of changeov

ines restart c

t additional

s’ competiti

rtunity costs [

pplication at

e main purpin Portugal

Fig

ement or po

sk that may

and adjustm

as illustrated

sented in Fig

ically reduce

mprovement

portant imp

changeover

onsiderable

novel ideas

hodology ha

his work can

plastic moul

Kara [22]. Ilysis of the ch

operations, iievement wa

le digit time

ead time at m

ribes an appli

eover time r

achines’ non

sful implem

eved. The firctivity incre

cognized tha

ver times. Tyig losses on

eovers. As a

[19]. Additio

(ii) improve

of producti

r time. With

n be avoided

enefits expe

eness is the

25].

GE Power

se of this waiming at im

re 1: Standardi

itioning of p

involve a nu

nts, have bee

in Figure 2,

Figur

ures 1 and 2

the changeov

of parts/tools

ct of imple

process carri

eduction in

nd solution c

been studie

e found in th

ing machine

this particulangeover pro

mprovements the reducti

value aimed

achines and a

cation case o

duction from

productive ti

ntation of t

t clear benefise [17] and t

t inventory l

ically, largeproduction c

consequence

ally, as poin

ent of the u

ity and (iv)

simple and e

. Quality and

ted from ad

substantial re

ontrols Port

rk was the iroving perf

3

ized height of d

rts and the s

ber of adjus

n developed.

hus eliminati

e 2: Centring ji

re two exam

er times of di

storage and

enting para

d out simult

hangeover ti

an be found i

and applied

e literature.

s, the descri

lar case, chancess. Special

of water con of change

by SMED, t

lso substantia

the SMED

30,4 to 9,68

e, which wa

he SMED

t is the reducto better com

evels are di

changeover tipacity. Obvi

batch sizes c

ed out by Shi

e of space a

elimination

ficient chang

safety impro

quate SME

duction in ch

ugal

plementatiormance of m

ie and clampin

tup of tools o

tments. To o

One commo

ng the need f

igs [7]

les of typica

es. Other pra

transportation

llel operation

aneously by

me whenever

Shingo’s re

by academici

tion of an i

geover teamattention has

nnections anver time fro

he reduction

l increase in

ethodology t

minutes was

s used in fav

ethodology

ion of unprodanies’ capabi

ectly related

mes, lead to lously, reducti

n be reduced

ingo [7], low

ea (less stora

of safety inv

eovers, error

ements and r

implementa

angeover cos

n of the SMetallic and p

g [7]

r dies in a m

ercome diffi

ly used in S

r adjustment

l solutions fr

tical techniq

and automat

s. Parallel o

ifferent oper

it can be im

erence book [

ans and peop

dustrial app

were organibeen given to

moulds’ cl365 to 73

was quite im

achine capa

o punch pres

attained. Thi

ur of the clie

everal imm

uctive time.lity to respon

to batch siz

arge batch sizon of change

substantially

r inventory l

ge area), (iii)

entory due t

are reduced

duction of p

ions [18]. P

ts, which inc

D methodollastic parts p

chine can be

ulties, novel

ED solution

s.

quently used

ues may inclu

ion [21]. It s

perations are

ators. This is

plemented.

7].

le from indus

lication of S

zed to collecthe separatio

lamping, tranminutes. Des

pressive ena

ity and utiliz

ses relevant t

s improveme

t.

diate benefi

his leads tod to changes

es which, in

es and smallover times all

, the same ha

vels allow: (i

reduction of

o changeove

and the trial

rocesses’ vari

rticularly be

lude labour,

gy at a powroduction and

a difficult

solutions,

s is a pair

in SMED

de the use

ould also

different

a strategy

Details of

try. Many

ED was

data andn between

sport andpite being

ling large

ation.

the work

t resulted

s for the

roductionin market

turn, are

number of ows more

pening to

) increase

inventory

mistakes

ests when

ability are

eficial to

quipment

r controlsreducing



inventormouldin

In

Video re

changeo

operators

lack of csaid to th

A

context o

to this.

represent

T

costs (

(1).

finally,annual sa

A

an organi

process

charts”.

were esti

A

to operat

a few id

below.

Punc

Incleaning

operation

S

to the e

machine

In

and, wh

replacedfaster.

T

improve

levers w

connecti

. Three specimachine.

order to clea

ordings wer

er procedure

to collect an

hangeover ine organizatio

s WIP reduct

f the ongoin

hus the ini

ing 80% of to

allow the q

) associated

represents

the machinvings per eac

nother releva

zed and opti

s little as po

ased on dir

mate.

fter the analy

ors’ involvem

as for impr

h-bending m

the punch-beand several

s.

me internal

istence of sp

are running.

terms of op

never possib

by bell crank

(a) (b)

e use of gau

some operati

as proposed

ns.

ific machines

rly understan

also made f

were not eff

look for die

structions anand location

ion in the ar

lean manufa

ial average

tal productio

antification

to the initial

he lost produ

e cost. The ah minute of r

t aspect in th

ized change

ssible. Opera

ct observatio

is of the initi

ent. From thi

vement cam

achine

nding machiarts’ preasse

perations inv

are sliding u

rations strea

le, the holes

s (Figure 3(b

Figure 3: (

es was intro

ns, not only i

and imple

were involv

d the change

r detailed an

icient. In addi

s, tooling and

workstationof tooling an

ea of metal

cturing impr

values for th

in the corres

f the moneta

cenario were

(1)

ction cost,

nnual changeduction in th

e assessment

ver, an opera

tors’ movem

n and analysi

al scenario, th

s and mainly

up. Some

e, internal anmbling are n

olving slidin

nits which a

lining, most

were modifi

)). Tools bec

) Use of screw

uced to elimi

n terms of ti

ented (Figu

4

ed: a punch-

ver processe

alysis and m

tion to high

other auxilia

s’ organizatid other elem

nd plastics

vement proj

is indicator

ponding area

ry gains obta

determined.

the labour c

over cost asschangeover

of changeove

tor should m

nts were sh

is of video r

e SMED met

following so

f the solutio

d external opow carried o

units were c

e now set-u

of the traditi

d for easy s

me unnecess

s with knob, (b

nate trial and

e spent but a

e 4(b)). Qu

ending mach

s, several cha

asuring of ch

hangeover ti

ry devices an

n could cleants associate

arts was one

cts, the SME

were also

s.

ined after SM

The tangible

ost, the sc

ciated to eactime was det

r efficiency i

ove out of his

wn and mea

cords, the tr

hodology wa

e brainstorm

ns and propo

erations wereut while the

onverted into

, i.e., assem

onal screws

etting (Figur

ary and the p

) Application o

error adjustm

lso regarding

ick coupling

ine, a punch

ngeover occu

angeover tim

es, high dist

documents.

ly be improvto changeov

of the com

D project sh

easured. Thi

ED impleme

osts were ca

rap cost,

h machine inrmined.

related to o

working are

sured using t

velled distan

applied. Spe

ing sessions

sals for impr

clearly separmachine is r

external ones

led with be

ere replaced

3(a)). In ot

rocess of tig

f bell cranks

ents, as illust

the ergonomi

s were adop

press and an

rrences were

es. It was cle

ances were tr

Moreover, th

ed. The samers.

any’s prioriti

uld strongly

s was based

ntation, the c

lculated usin

the inventor

volved and t

erators’ mov

during the c

he so-called

ces during c

cial attention

ith the opera

ovement are

ated. Lookinunning, i.e. a

. This was po

rings and ca

by screws

her cases sc

tening/loose

rated in Figu

ic conditions,

ted for the

injection

observed.

ar that the

avelled by

re were a

could be

es, in the

contribute

on parts

hangeover

Equation

cost and,

e average

ments. In

angeover

“spaghetti

angeover

was given

tors, quite

described

for tools,s external

ssible due

ms, while

ith knobs

ews were

ing much

e 4(a). To

the use of

hydraulic



Di

delays di

T

improve

standardi

departme

O

changeo

Punc

Fmachine,

particula

operation

the mac

machine.

T

adjustme

in the pr

carry out

opening/

P

possible

there we

ies exchange

s exchange c

e standardiz

ents in the

zation of sup

nt of the com

her improve

er process of

h press

llowing theinternal and

, a small su

s into extern

ine, with th

streamline

nts, both in th

ss was impl

the centring

losing of die

rallel operati

for two main

e enough ope

(a)

Fig

at the machi

onsiderably.

tion of supp

hangeover p

ort plates wa

pany.

ments, inclu

the punch-be

SMED stratexternal ope

pport plate

l ones. When

machine st

he changeov

e press and i

mented. Thi

operation.

s, Figure 6(a)

(a)

Figure 6:

ons were als

reasons. First

rators availab

re 4: (a) Appli

nes were aid

igure 5: Non-

ort plates for

rocess and s

s made to the

ding the int

nding machin

gy for imprrations were

or a countin

a single sup

pped, and fi

r operations

a raw materi

made the ce

oreover, the

Figure 6(b) s

(a) Use of hyd

o implement

the changeo

le to carry ou

5

cation of gaug

ed by non-s

tandard suppo

dies was re

eeding up t

company. T

roduction of

e.

oving changidentified fo

sensor wa

ort was avai

xed in the n

the height o

ial feeder. Ad

ntring much

application

hows the use

raulic cylinder

d in the cha

er process o

t the parallel

s, (b) Use of l

andard supp

rt plates for di

ognized as

e exchange

is is currentl

parallel op

eovers, as alr the change

duplicated

lable this had

ew die, whi

f dies was st

ditionally a p

faster since

f hydraulic c

of a typical s

s (b) Tools sha

ngeover proc

erations coul

operations str

(b)

vers

rt plates (Fi

s

ery suitable

of dies. Due

being dealt

ration, wer

ready donever process

or allowing

to be remov

h would rep

ndardized, e

oka-yoke sys

reviously a

ylinders allo

hadow board

(b)

ow board

esses of the

physically a

tegy

ure 5). This

solution for

to this, a pr

with by the e

implement

for the puncof the punch

conversion

d from the di

lace the for

liminating th

tem for centri

icrometre w

ed a faster

to organize th

punch press.

llow it; secon

approach

ubstantial

posal for

gineering

d in the

h-bendingpress. In

f internal

e setup in

er in the

us several

ng the die

as used to

nd easier

e tools.

This was

d because



InjecIn

tooling a

machine

operation

die to be

and inserwas runn

previousl

C

8(b), tool

machine’

and closi

In

like 5S [

Results

A

having i

necessar

impleme

indicatora change

ion mouldinthe injection

nd auxiliary

is running.

s required st

changed. Thu

ting it, whening, i.e., as a

y performed i

ncerning the

ls shadow bo

s control pan

ng the die, pr

general term

6] were appli

nd discussiofter SMED i

to considera

data was g

ted. As pre

s were considover. Table 1

g machinemoulding m

set-up devic

dditionally,

pping the m

s, the first co

needed, inton external op

nside the mo

Figu

streamline o

rds were app

el. These par

ssure and oth

igure 8: (a) Se

s, checklists a

ed in order t

nprovement

tion the initi

athered befo

iously referr

ered, namelyshows the at

chine, like in

s needed to

two internal

chine and re

nversion was

he die to beeration. The

lding machi

e 7: Connector

changeover

lied and a set

ameters inclu

ers.

curity clampin

nd changeov

improve wor

proposals we

al values of

re, as above

ed, in additi

WIP, annualtained impro

6

the other ma

quickly perf

operations

oving the si

attained sim

sed next. Thisecond conv

e and is now

between clam

operations, cl

of pre-establi

ded temperat

(a)

(b)

system for tr

r instruction

kstations’ or

re implemen

the perform

referred, an

n to change

changeover cements on th

chines, all pr

orm the chan

ere converte

ngle availabl

ly by manuf

is assembly crsion involve

executed out

ing cylinder a

lamping syste

ished values

ure, amount

nsport, (b) die

were created

anization.

ted, their im

nce measure

after the S

over times, o

osts and distae changeove

paratory wor

geover, is n

d into extern

connector to

cturing anoth

uld then bed pre-heating

ide.

d die

ms were imp

as defined f

f material in

lamping syste

for all the m

act on perf

s selected an

MED impro

ther three qu

nce travelledtime, for ea

k, for making

ow executed

al ones. On

be inserted i

er connector

one while thof the die,

roved, Figure

or the parame

ected, time

m

achines, and t

rmance was

d above ref

ement propo

antitative pe

by the operath machine, f

ready the

while the

of these

n the next

(Figure 7)

machinehich was

s 8(a) and

ters in the

f opening

echniques

evaluated

rred. The

sals were

rformance

ors duringor each of



7

three stages of evolution of the SMED study. The improvements results are, in general, impressive, i.e., 58,7,

89,7 and 83,7 % improvement respectively for the Punch-Bending, Punch Press and Injection, moulding

machines. Nevertheless, in the case of punch-bending machine the solutions adopted didn’t achieve the single

digit value for changeovers as aimed at by SMED. This simply means that there is still scope for improvement.In the other two machines the objective was accomplished.

Table 1. Improvements on changeover time for the three improvement stages.

Machines

Changeover time (min) Overall

Changeover time

improvement (%)Initial 1st stage 2nd stage 3rd stage

Punch-Bending Machine 188,15 146,02 108,00 77,80 58,7

Punch Press 61,80 54,00 52,33 6,35 89,7

Injection Moulding Machine 58,14 54,48 17,72 9,80 83,1

In addition to the improvements of changeover times, productivity, workstations ergonomic conditions

and work organization were considerably improved too. In particular a clear and systematic procedure was

established for each changeover in each machine, supported by checklists strictly followed by operators.

As expected, the reduction of batch sizes allowed by the decrease of changeover times has led to a

reduction of WIP. Table 2 compares WIP values for metallic and plastic parts, before and after SMED study and

implementation. The comparison reveals very significant WIP reductions, i.e., approximately 55 and 80% forboth time and money, in the areas of metal and plastic components, respectively.

Table 2. Impact of the SMED solutions on WIP: Comparing before with after performance

Production area

Average WIP (days) Average WIP (€)

Before AfterImprove-

ment (%)

Before After Improve-

ment (%)

Metal parts 17,05 7,74 54,6 318,75 144,38 54,7

Plastic parts 5,00 1,09 78,2 439,00 87,80 80,0

Changeover costs were also evaluated. Calculations were based on Equation (1). Results are shown in

Table 3.

Table 3. Measuring costs and savings from SMED implementation

Machines

Average annual changeover cost(€)

Before AfterImprove-

ment (%)

Savings

(€)

Punch-Bending Machine 19.370,9 8.009,9 58,7 11.361,1

Punch Press 5.076,4 521,6 89,7 4.554,8

Injection Moulding Machine 4.840,0 815,8 83,1 4.024,2

Total 19.940,1

The reduction of operators’ movements was one of the important aspects contributing to the changeovers

improvement. This was achieved by external operations concerned with placing near the machines all the

necessary tools, devices, materials, documents and instructions. Table 4 shows the new values for the operators’travelled distance during a changeover occurrence.

Table 4. Comparing operators’ movements before and after SMED implementation

MachinesTravelled distance (m) Savings

(%)Before After

Punch-Bending Machine 370 10 97,3

Punch Press 260 2 99,2

Injection Moulding Machine 300 10 96,7

When compared with the initial distance travelled values these results from the SMED study show

dramatic savings, i.e., 97,3, 99,2 and 96,7%, respectively for the punch-bending machine, punch press and

injection moulding machine.



8

ConclusionsKeeping up with competition in the economic global market of today, characterized by increasing product

customization, requires efficient use of resources and customer service effectiveness. In manufacturing this is

highly dependent on product flexibility. This means the ability to change production from one product to anotherwithout relevant additional costs. This contributes to both, customer service effectiveness, by providing quick

response to demand, and to production efficiency, at several dimensions. However product flexibility should not

be a burden on production capacity. To manage this SMED technology can give a fundamental contribution.This involves a methodology and a number of techniques and changeover strategies and solutions for quick

product changeover.

This paper describes the application of SMED technology in a power controls industrial company of the

north of Portugal in the production areas of metal and plastic components. After a brief literature review, the

paper puts forward important strategies and techniques of SMED, emphasizing the importance of converting

internal into external operations. Additionally two important strategies of high impact on reducing changeover

times, namely quick centring and adjustment of dies at machines, for fast changeovers, and the use of parallel

changeover operations, are briefly described.

The industrial case studied is described in line with most important SMED strategies and novel solutions.

Many proposals were made with visible and impressive impact on changeover processes, in terms of time and

cost savings, reduction of work-in-process levels and distance travelled by operators during the changeover

process. Most of the proposals were implemented and the real impact of solutions assessed and reported. Quick

changeovers allowed reducing batch sizes and increasing product flexibility, due to the increased number of product changeovers without burdening production capacity. A much smoothed production flow and smaller

lead times were obtained.

In terms of changeover time, reductions ranging from 58 to almost 90% were achieved. The WIP of metal

components was reduced from 17.05 to 7.74 days and from 5 to 1.09 days in the case of plastic components. The

distance travelled by operators during changeover processes was dramatically reduced from 370 to 10 m (punch

bending machine), 260 to 2m (punch press) and 300 to 10m (injection moulding machine).

The involvement of operators, as advocated by SMED, proved to be critical in the success of the studyand its implementation. In addition to operators’ participation in the implementation of solutions their

contribution was also important for generating ideas for good solutions.

Although large benefits were obtained from this project, scope for further improvement still exists. In fact

the objective of changeover times below 10 minutes aimed by SMED was not achieved for the punch-bending

machine.

This project has clearly shown the potential of SMED to improve manufacturing efficiency andeffectiveness of industrial companies.

AcknowledgmentsThis work had the financial support of FCT-Fundação para a Ciência e Tecnologia of Portugal under the

project PEst-OE/EME/UI0252/2011. We also acknowledge the institutional support given by GE Power Controls

Portugal and University of Minho.

Bibliography[1] Duguay C. R., Landry S., Pasin F., “From Mass Production to Flexible/Agile Production”. Journal of

Operations and Production Management. 1997; 17, 12, 1183-1195

[2] Department of Trade and Industry (DTI), 1994, “Competitive Manufacturing – A Practical Approach to the

Development of a Manufacturing Strategy”. DTI, London. 1994.

[3] Womack J. P., Jones D. T. and Roos D., “The machine that changes the world”. Rawson Associates, NY.1990.

[4] Womack J. P., Jones D. T., “Lean Thinking: Banish Waste and Create Wealth in Your Corporation”. Free

Press. 1996.

[5] Ohno T., “Toyota Production System: Beyond Large-Scale Production”, Productivity Press. 1988.[6] Monden Y., “Toyota Production System”. 1st edition, Industrial Engineering and Management Press. 1988.

[7] Shingo S., “Revolution in Manufacturing: Single-minute Exchange of Die System”. Productivity Press.1985.

[8] J. Carlos Sá, J. Dinis-Carvalho and Rui M. Sousa (2011) “Waste Identification Diagrams”, in 6º Congresso

Luso-Moçambicano de Engenharia (CLME'2011), Maputo, Mozambique. 2011.

[9] Rother M., Shook J., “Learning to See: Value Stream Mapping to Add Value and Eliminate Muda”. The

Lean Enterprise Institute. 1999.

[10] Karlsoon C., Ahlstrom P., “Assessing Changes toward Lean Production”. Journal of Operations and

Production Management. 1996: 16, 2, 24-41.[11] Sekine K, Arai K., “Kaizen for Quick Changeover: Going Beyond SMED”. Productivity Press, 1992.



9

[12] Prasad B., “A structured methodology to implement judiciously the right JIT tactics”. Production Planning

& Control. 1995; 6, 6, 564-577.

[13] Venjara Y., “Setup Savings”. Manufacturing Engineering. 1996; 117, Jul., 96-102.

[14] Van Goubergen D., “Set-up Reduction as an Organization-wide Problem”. Proceedings of the IIE Solutions.2000.

[15] Spence A. M., Porteus E. L., “Setup Reduction and Increased Effective Capacity”. Management Science.

1987: Vol 33, 1291-1301.[16] Fernihough A, Culley S.J., Mileham A.R., “A Financial Model for Setup Reduction Justification”.

Proceedings of the Fourth International FAIM Conference, Blacksburg, 1994.

[17] Yang J., Deane R., “Setup Time Reduction and Competitive Advantage in a Closed Manufacturing Cell”.

European Journal of Operation Research. 1993; 69, 3, 413-423.

[18] Hay E.J., “Any Machine Setup Time Can be Reduced by 75%”. Industrial Engineering, Volume 19, 1987.[19] Shingo S., “Quick Changeover for Operators: The SMED System”. Shopfloor Series, Productivity Press,

1996.

[20] Sousa R. M., Lima R. M., Carvalho D. and Alves A., “An Industrial Application of Resource Constrained

Scheduling for Quick Changeover”. In 2009 IEEE International Conference on Industrial Engineering and

Engineering Management, H. Sun, R. Jiao & M. Xie (Editors), 189-193, Hong Kong, China. 2009.

[21] Termini M. J., “The New Manufacturing Engineer”. Society of Manufacturing Engineers. 1996.

[22] Kays B., Kara S., “Set-Up Reduction in Injection Molding Process – A Case Study in Packaging Industry”.

4th International Conference and Exhibition on Design and Production of Machines and Dies/Molds. 2007.[23] Brian M. T., “Application of Shingo's Single Minute Exchange of Dies (SMED) Methodology to Reduce

Punch Press Changeover Times at Krueger International”. University of Wisconsin-Stout. 2007.

[24] X. A., Vonderembse M. A., “The Impact of organizational Structure on the Level of JIT Attainment:

Toward Theory Development”. International Journal of Production Research. 1998; 36, 10, 2873-2878

[25] Productivity Press, “Improving Flow: Collected Practices and Cases – Insights on Implementation”.

Productivity Press, 1st edition. 2006.

[26] Tapping, D., “The Lean Office Pocket Guide XL”, MCS Media, Inc.. 2005.

dr-fb-rs-cs rrpmom 40 (2011) pre-print

Documents