Mah o, Dalgobind; Kuma , Anjani
A icle
Applica ion o oo cause analysis in imp o emen o
p oduc quali y and p oduc i i y
Jou nal o Indus ial Enginee ing and Managemen (JIEM)
P o ided in Coope a ion wi h:
The School o Indus ial, Ae ospace and Audio isual Enginee ing o Te assa (ESEIAAT), Uni e si a
Poli ècnica de Ca alunya (UPC)
Sugges ed Ci a ion: Mah o, Dalgobind; Kuma , Anjani (2008) : Applica ion o oo cause analysis
in imp o emen o p oduc quali y and p oduc i i y, Jou nal o Indus ial Enginee ing and
Managemen (JIEM), ISSN 2013-0953, OmniaScience, Ba celona, Vol. 1, Iss. 2, pp. 16-53,
h ps://doi.o g/10.3926/jiem. 1n2.p16-53
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/188371
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by-nc/3.0/
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 16
D. Mah o; A. Kuma
Applica ion o oo cause analysis in imp o emen o p oduc
quali y and p oduc i i y
Dalgobind Mah o; Anjani Kuma
Na ional Ins i u e o Technology (INDIA)
[email p o ec ed]; [email p o ec ed]
Recei ed July 2008
Accep ed Decembe 2008
Abs ac
: Roo -cause iden i ica ion o quali y and p oduc i i y ela ed p oblems a e key
issues o manu ac u ing p ocesses. I has been a e y challenging enginee ing p oblem
pa icula ly in a mul is age manu ac u ing, whe e maximum numbe o p ocesses and
ac i i ies a e pe o med. Howe e , i may also be implemen ed wi h ease in each and e e y
indi idual se up and ac i i ies in any manu ac u ing p ocess. In his pape , oo -cause
iden i ica ion me hodology has been adop ed o elimina e he dimensional de ec s in
cu ing ope a ion in CNC oxy lame cu ing machine and a ejec ion has been educed
om 11.87% o 1.92% on an a e age. A de ailed expe imen al s udy has illus a ed he
e ec i eness o he p oposed me hodology.
Keywo ds:
oo cause analysis, cause and e ec diag am, in e ela ionship diag am and
cu en eali y ee
1. In oduc ion
In Roo Cause Analysis (RCA) is he p ocess o iden i ying causal ac o s using a
s uc u ed app oach wi h echniques designed o p o ide a ocus o iden i ying and
esol ing p oblems. Tools ha assis g oups o indi iduals in iden i ying he oo
causes o p oblems a e known as oo cause analysis ools. E e y equipmen ailu e
happens o a numbe o easons. The e is a de ini e p og ession o ac ions and
consequences ha lead o a ailu e. Roo Cause Analysis is a s ep-by-s ep me hod
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 17
D. Mah o; A. Kuma
ha leads o he disco e y o aul s o oo cause. An RCA in es iga ion aces he
cause and e ec ail om he end ailu e back o he oo cause. I is much like a
de ec i e sol ing a c ime.
To mee up he high changing ma ke demands along wi h high quali y a
compa able p ices, one shall ha e o iden i y quickly he oo causes o quali y
ela ed p oblems by e iewing an e en , wi h he goals o de e mining wha has
happened, why i has happened and wha can be done o educe he likelihood o
ecu ence.
2. Objec i e and ou line o he s udy
The e a e a ie ies o p oblems ela ed o p oduc quali y and p oduc i i y in
indus ies due o a ying deg ees o abno mali y and ine iciency which ul ima ely
causes ejec ion. Roo -cause iden i ica ion o quali y- ela ed p oblems is a key and
necessa y s ep in he ope a ions o manu ac u ing p ocesses, especially in high-
h oughpu au oma ed p ocesses.
This is p edominan ly ue o he mul is age manu ac u ing p ocesses, which is
de ined as a p ocess ha p oduces he p oduc s unde mul iple se ups. The quali y
in o ma ion low o he p oduc in a mul is age manu ac u ing sys em and he
in e ac ion be ween he p ocess aul s and he p oduc quali y cha ac e is ics a e
e y complica ed. In mul is age p ocess, he iden i ica ion o p ocess oo cause is
also no simple. I has been obse ed ha he implemen a ion o Roo Cause
Analysis in a pa icula single indi idual se up has simpli ied he p oblem.
A case s udy was done o an indus y which was in dold ums condi ion. The ab up
shu downs and b eakdowns (5.19% o annual sales), equen cus ome complain s
(367pa), line balancing delay (27%), ma e ial sca ci y o una ailabili y o ma ching
ma e ial (58 days pa), ejec ion (3.03% o sales) and a ious o he key success
ac o s we e no up o he ma k. The e o e, he Roo Cause Analysis was
unde aken o imp o e he plan si ua ion. Bu , he s udy was con ined o he CNC
Oxy Flame Cu ing Machine. In his pape , he iden i ica ion o he p oblem has
been simpli ied aking in o conside a ion a pa icula s age o manu ac u ing. I has
been obse ed ha RCA can also be implemen ed in each and e e y indi idual se
up o manu ac u ing o imp o e p oduc quali y and p oduc i i y.
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 18
D. Mah o; A. Kuma
3. Li e a u e e iew
Wilson e al. (1993) ha e de ined he Roo Cause Analysis as an analy ic ool ha
can be used o pe o m a comp ehensi e, sys em-based e iew o c i ical inciden s.
I includes he iden i ica ion o he oo and con ibu o y ac o s, de e mina ion o
isk educ ion s a egies, and de elopmen o ac ion plans along wi h measu emen
s a egies o e alua e he e ec i eness o he plans.
Canadian Roo Cause Analysis F amewo k (2005) says ha oo cause analysis is
an impo an componen o a ho ough unde s anding o “wha happened”. The
eam begins by e iewing an “ini ial unde s anding” o he e en and iden i ying
unanswe ed ques ions and in o ma ion gaps. The in o ma ion-ga he ing p ocess
includes in e iews wi h s a , who we e di ec ly and indi ec ly in ol ed,
examina ion o he physical en i onmen whe e he e en and o he ele an
p ocesses ook place, and obse a ion o usual wo k p ocesses. This in o ma ion is
syn hesized in o a “ inal unde s anding”, which is hen used by he eam o begin
he “why” po ion o he analysis.
Simila ly, o sol e a p oblem, one mus i s ecognize and unde s and wha is
causing he p oblem. This is he essence o oo cause analysis. Acco ding o
Wilson e al. (1993) a oo cause is he mos basic eason o an undesi able
condi ion o p oblem. I he eal cause o he p oblem is no iden i ied, hen one is
me ely add essing he symp oms and he p oblem will con inue o exis .
Dew (1991) and Sp oull (2001) s a e ha iden i ying and elimina ing oo causes o
any p oblem is o u mos impo ance. Roo cause analysis is he p ocess o
iden i ying causal ac o s using a s uc u ed app oach wi h echniques designed o
p o ide a ocus o iden i ying and esol ing p oblems. Tools ha assis g oups and
indi iduals in iden i ying he oo causes o p oblems a e known as oo cause
analysis ools.
Acco ding o Dugge (2004) se e al oo cause analysis ools ha e eme ged om
he li e a u e as gene ic s anda ds o iden i ying oo causes. Some o hem a e
he Why Why Analysis, Mul i Va i Analysis, Cause-and-E ec Diag am (CED), he
In e ela ionship Diag am (ID), and he Cu en Reali y T ee (CRT). He has added
ha Why Why analysis is he mos simplis ic oo cause analysis ool whe e as
cu en eali y ee is used o possible ailu es o a sys em and i is commonly
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 19
D. Mah o; A. Kuma
used in he design s ages o a p ojec and wo ks well o iden i y causal
ela ionships. The e is no sho age o in o ma ion a ailable abou hese ools.
The li e a u es con i med ha hese ools do, in ac , ha e he capaci y o ind he
oo causes wi h a ying deg ees o accu acy, e iciency, and quali y. DOE
Guideline Roo Cause Analysis Guidance Documen Feb ua y (1992) says ha
immedia ely a e he occu ence iden i ica ion, i is impo an o begin he da a
collec ion phase o he oo cause p ocess using hese ools o ensu e ha da a a e
no los . The da a should be collec ed e en du ing an occu ence wi hou
comp omising wi h sa e y o eco e y. The in o ma ion ha should be collec ed
consis s o condi ions be o e, du ing, and a e he occu ence; pe sonnel
in ol emen (including ac ions aken); en i onmen al ac o s; and o he
in o ma ion ha ing ele ance o he condi ion o p oblem. Fo se ious cases,
pho og aphing he a ea o he occu ence om se e al iews may be use ul in
analysis. E e y e o should be made o p ese e physical e idence such as ailed
componen s, up u ed gaske s, bu ned leads, blown uses, spilled luids, and
pa ially comple ed wo k o de s and p ocedu es. This should be done despi e
ope a ional p essu es o es o e equipmen o se ice. Occu ence pa icipan s and
o he knowledgeable indi iduals should be iden i ied.
Ande son and Fage haug (2000) ha e simpli ied he oo cause analysis. They
p o ide a comp ehensi e s udy abou he heo y and applica ion o me ics in oo
cause analysis. I emphasizes he di icul y in achie ing p ocess capabili y in
so wa e domain and is cau ious abou SPC implemen a ion. They men ion ha he
use o con ol cha s can be help ul o an o ganiza ion especially as a
supplemen a y ool o quali y enginee ing models such as de ec models and
eliabili y models. Howe e , i is no possible o p o ide con ol as in manu ac u ing
since he pa ame e s being cha ed a e usually in-p ocess measu es ins ead o
ep esen ing he inal p oduc quali y. The inal p oduc quali y can only be
measu ed a he end o a p ojec as opposed o he p oduc ion in manu ac u ing
indus y, so ha on- ime con ol on p ocesses becomes impossible. They also
unde line he necessi y o ma u i y o achie ing p ocess s abili y in de elopmen
o p oduc quali y and p oduc i i y. Finally, hey b ing a elaxed unde s anding by
s a ing ha he p ocesses can be ega ded in con ol when he p ojec mee s in-
p ocess a ge s and achie es end-p oduc quali y and p oduc i i y imp o emen
goals.

doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 20
D. Mah o; A. Kuma
A ca o (1997) has p esen ed a ious ools o iden i ying oo causes. He desc ibes
ha RCA echniques a e cons ained wi hin domain and gi e a de ailed u o ial by
suppo ing heo e ical knowledge wi h p ac ical expe iences. He s a es ha all RCA
echniques may no be applicable o all p ocesses.
B own (1994) has used he oo cause echnique o analyze he assembly o
comme cial ai c a . He has concluded ha i is he mos e ec i e ool o elimina e
he causes in mos i al assemblies like ai c a , whe e u mos sa e y and eliabili y
is needed.
B assa d (1996), and B assa d and Ri e (1994) ha e pu hei emphasis on
con inuous imp o emen and e ec i e planning. They ha e poin ed ou ha Roo
Cause analyzing ools gi e managemen o hink ahead abou ailu es and plan
acco dingly. They emphasize ha p ocess imp o emen models implici ly di ec
companies o implemen RCA as a c ucial s ep o p ojec le el p ocess con ol and
o ganiza ional le el p ocess imp o emen pu poses. Quan i a i e P ocess
Managemen equi es es ablishing goals o he pe o mance o he p ojec 's
de ined p ocess, aking measu emen s o he p ocess pe o mance, analyzing hese
measu emen s, and making adjus men s o main ain p ocess pe o mance wi hin
accep able limi s.
Cox and Spence (1998) ha e ad oca ed ha RCA ools e ec i ely gi e solu ion o
handle cons ain s and a i e a an app op ia e decision. Like Cox and Spence
(1998), De me (1997) has also used oo cause analysis on managemen o
cons ain s. He p esen s one o he ea lies s udies on he deba e o applying Roo
Cause Analysis o p ocesses. A p ope managemen decision is necessa y o
succeed he RCA ools and me hods in a pa icula en i onmen .
Lepo e and Cohen (1999), Mo an e al. (1990), Robson (1993) and Scheinkop
(1999) mo e ahead ha when change is needed, hen hink oo cause analyzing,
iden i ying and elimina ing. The ounda ions o hei s udies a e pionee ing one as
hey ques ion an accep ed p ac ice o oo cause analysis and he esul s o he
example s udies a e encou aging. Howe e , he s udies a e a om being p ac ical
one as hey include oo many pa ame e s and assump ions.
Smi h (2000) has explained ha Roo Cause Tools can esol e con lic ing
s a egies, policies, and measu es. The pe cep ion is ha one ool is as good as
ano he ool. While he li e a u e was qui e comple e on each ool as a s and-alone
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 21
D. Mah o; A. Kuma
applica ion and hei ela ionship wi h o he p oblem sol ing me hods. The e a e
e y ew li e a u es a ailable on he compa a i e s udy o a ious oo cause
analysis ools and me hods. The s udy on h ee ools namely Cause-and-E ec
Diag am (CED), he In e ela ionship Diag am (ID), and he Cu en Reali y T ee
(CRT) is de icien on how hese h ee ools di ec ly compa e o each o he . In ac ,
he e a e only wo s udies ha compa ed hem and he compa isons we e
quali a i e.
Likewise, F edendall e al. (2002) ha e also compa ed he CED and he CRT using
p e iously published examples o hei sepa a e e ec i eness. While Pasqua ella e
al. (1997) compa ed CED, ID and CRT on Equipmen /Ma e ial P oblem, P ocedu e
P oblem, Pe sonnel E o , Design P oblem, T aining De iciency, Managemen
P oblem and Ex e nal Phenomena using a one-g oup pos - es design wi h
quali a i e esponses.
The e is li le published esea ch ha quan i a i ely measu es and compa es he
Why Why Analysis, Mul i Va i Analysis, Cause-and-E ec Diag am (CED), he
In e ela ionship Diag am (ID), and he Cu en Reali y T ee (CRT).
Geno (2007) has p esen ed some insigh in o he compa ison o common oo
cause analysis ools and me hods. He indica es ha he e a e some compa a i e
di e ences be ween ool and me hod o a RCA. He has added ha ools a e
included along wi h me hods because ools a e o en ou ed and used as a ull-
blown oo cause analysis.
4. Basic e minologies in oo cause analysis
 Facili y: Facili y may be de ined as any equipmen , s uc u e, sys em,
p ocess, o ac i i y ha ul ills a speci ic pu pose. Some o he examples
include p oduc ion o p ocessing plan s, accele a o s, s o age a eas, usion
esea ch de ices, nuclea eac o s, coal con e sion plan s, magne o
hyd odynamics expe imen s, windmills, adioac i e was e, disposal sys ems,
es ing and esea ch labo a o ies, anspo a ion ac i i ies, and
accommoda ions o analy ical examina ions o i adia ed and unp edic ed
componen s.
 Condi ion: I may be de ined as a s a e, whe he o no esul ing om an
e en , ha may ha e ad e se sa e y, heal h, quali y assu ance, secu i y,
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 22
D. Mah o; A. Kuma
ope a ional, o en i onmen al implica ions. A endi ion is usually
p og amma ic in na u e; o example, an (exis ing) e o in analysis o
calcula ion, an anomaly associa ed wi h ( esul ing om) design o
pe o mance, o an i em indica ing weaknesses in he managemen p ocess
a e all condi ions.
 Roo Cause: The cause ha , i co ec ed, would p e en ecu ence o his
and simila occu ences. The oo cause does no apply o his occu ence
only, bu has gene ic implica ions o a b oad g oup o possible occu ences,
and i is he mos undamen al aspec o he cause ha can logically be
iden i ied and co ec ed. The e may be a se ies o causes ha can be
iden i ied, one leading o ano he . This se ies should be pu sued un il he
undamen al, co ec able cause has been iden i ied. Fo example, in he
case o a leak, he oo cause could be managemen , no i s main enance,
which ensu es ha i is e ec i ely managed and con olled. This cause
could ha e led o he use o imp ope seal ma e ial o missed p e en i e
main enance on a componen , which ul ima ely led o he leak. In he case
o a sys em misalignmen , he oo cause could be a p oblem in he aining
p og am, leading o a si ua ion in which ope a o s a e no ully amilia wi h
con ol oom p ocedu es and a e willing o accep excessi e dis ac ions.
 Causal Fac o : A condi ion o an e en ha esul s in an e ec (any hing
ha shapes o in luences he ou come). This may be any hing om noise in
an ins umen channel, a pipe b eak, an ope a o e o , o a weakness o
de iciency in managemen o adminis a ion. In he con ex o DOE he e
a e se en majo causal ac o ca ego ies. These majo ca ego ies a e:
o Equipmen /Ma e ial P oblem
o P ocedu e P oblem
o Pe sonnel E o
o Design P oblem
o T aining De iciency
o Managemen P oblem
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 23
D. Mah o; A. Kuma
o Ex e nal Phenomenon
5. Roo cause analysis ools and echniques
Many Roo Cause Analysis Tools ha e eme ged om he li e a u e as gene ic
s anda ds o iden i ying oo causes. They a e he Cause-and-E ec Diag am
(CED), he In e ela ionship Diag am (ID), and he Cu en Reali y T ee (CRT),
Why Why Analysis, Mul i Va i Analysis. Ample o in o ma ion is a ailable abou
hese ools, in open li e a u e (See Re e ences).
5.1 Causes-and-e ec diag am (CED)
This diag am, also called Ishikawa o Fishbone Diag am, is used o associa e
mul iple possible causes wi h a single e ec . The diag am is cons uc ed o iden i y
and o ganize he possible causes o a pa icula single e ec . Causes in Cause and
E ec Diag am a e equen ly a anged in ou majo ca ego ies. Fo manu ac u ing
cases i is Manpowe , Me hods, Ma e ials and Machine y. Fo Adminis a ion and
se ice sec o s, i is Equipmen , Policies, P ocedu es and People. Ishikawa
ad oca ed he CED as a ool o b eaking down po en ial causes in o mo e de ailed
ca ego ies so ha hey can be o ganized and ela ed in o ac o s which help in
iden i ying he oo cause.
5.2 In e ela ionship diag am (ID)
Mizuno suppo ed he ID as a ool o quan i y he ela ionships be ween ac o s and
he eby classi y po en ial causal issues o d i e s. The in e ela ionships among he
ope a ions a e shown as ‘in and ou ’ in each s ages o ope a ion. The weigh
ac o s, which may include causes, e ec s, o bo h, o in and ou a e de e mined
on he basis on logical sequence.
5.3 Cu en eali y ee (CRT)
Cu en Reali y T ee is a ool o ind logical in e dependen chains o ela ionships
be ween undesi able e ec s leading o he iden i ica ion o he co e cause. I
depic s he eal s a us unde p e ailing cu en condi ions wi h ega d o causali y,
ac o ela ionships, usabili y, and pa icipa ion.
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 30
D. Mah o; A. Kuma
 Wha would you ha e done di e en ly o ha e p e en ed he occu ence,
dis ega ding all economic conside a ions (as ega ds ope a ion,
main enance, and design)?
 Wha would you ha e done di e en ly o ha e p e en ed he occu ence,
conside ing all economic conce ns (as ega ds ope a ion, main enance and
design)?
6.4 Managemen o e sigh and isk ee (MORT) analysis
I is he me hodology adop ed by he oo cause analysis eam wi h he ac i e
suppo o managemen . To pe o m he MORT analysis:
 Iden i y he p oblem associa ed wi h he occu ence and lis i as he op
e en .
 Iden i y he elemen s on he "wha " side o he ee ha desc ibe wha
happened in he occu ence.
 Fo each ba ie o con ol p oblem, iden i y he managemen elemen s on
he "why" side o he ee ha pe mi ed he ba ie con ol p oblem.
 Desc ibe each o he iden i ied inadequa e elemen s o p oblems and
summa ize you indings.
A b ie explana ion o he "wha " and "why" may assis in using mini-MORT o
causal analyses.
 Ba ie s ha su ound he haza d and/o he a ge and p e en con ac o
con ols and p ocedu es ha ensu e sepa a ion o he haza d om he
a ge
 Plans and p ocedu es ha a oid con lic ing condi ions and p e en
p og amma ic impac s.
 In a acili y, wha unc ions implemen and main ain hese ba ie s,
con ols, plans, and p ocedu es?

doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 31
D. Mah o; A. Kuma
 Iden i ying he haza ds, a ge s, and po en ial con ac s o in e ac ions and
speci ying he ba ie s/con ols ha minimize he likelihood and
consequences o hese con ac s
 Iden i ying po en ial con lic s/p oblems in a eas such as ope a ions,
scheduling, o quali y and speci ying managemen policy, plans, and
p og ams ha minimize he likelihood and consequences o hese ad e se
occu ences
 P o iding he physical ba ie s: designing, ins alla ion, signs/wa nings,
aining o p ocedu es
 P o iding planning/scheduling, adminis a i e con ols, esou ces, o
cons ain s. Ve i ying ha he ba ie s/con ols ha e been implemen ed and
a e being main ained by ope a ional eadiness, inspec ions, audi s,
main enance, and con igu a ion/change con ol
 Ve i ying ha planning, scheduling, and adminis a i e con ols ha e been
implemen ed and a e adequa e
 Policy and policy implemen a ion (iden i ica ion o equi emen s, assignmen
o esponsibili y, alloca ion o esponsibili y, accoun abili y, igo and
example in leade ship and planning).
6.5 Human pe o mance e alua ion
Human Pe o mance E alua ion is used o iden i y ac o s ha in luence ask
pe o mance. I is mos equen ly used o man-machine in e ace s udies. I s
ocus is on ope abili y and wo k en i onmen , a he han aining ope a o s o
compensa e o bad condi ions. Also, human pe o mance e alua ion may be used
o mos occu ences since many condi ions and si ua ions leading o an occu ence
ul ima ely esul om some ask pe o mance p oblem such as planning,
scheduling, ask assignmen analysis, main enance, and inspec ions. T aining in
e gonomics and human ac o s is needed o pe o m adequa e human pe o mance
e alua ions, especially in man-machine in e ace si ua ions.
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 32
D. Mah o; A. Kuma
6.6 Kepne -T egoe p oblem sol ing and decision making
Kepne -T egoe is used when a comp ehensi e analysis is needed o all phases o
he occu ence in es iga ion p ocess. I s s eng h lies in p o iding an e icien ,
sys ema ic amewo k o ga he ing, o ganizing and e alua ing in o ma ion and
consis s o ou basic s eps:
 Si ua ion app aisal o iden i y conce ns, se p io i ies, and plan he nex
s eps.
 P oblem analysis o p ecisely desc ibe he p oblem, iden i y and e alua e
he causes and con i m he ue cause. (This s ep is simila o change
analysis).
 Decision analysis o cla i y pu pose, e alua e al e na i es, and assess he
isks o each op ion and o make a inal decision.
 Po en ial p oblem analysis o iden i y sa e y deg ada ion ha migh be
in oduced by he co ec i e ac ion, iden i y he likely causes o hose
p oblems, ake p e en i e ac ion and plan con ingen ac ion. This inal s ep
p o ides assu ance ha he sa e y o no o he sys em is deg aded by
changes in oduced by p oposed co ec i e ac ions.
These ou s eps co e all phases o he occu ence in es iga ion p ocess and hus,
Kepne -T egoe can be used o mo e han causal ac o analysis. This sys ems
app oach p e en s o e looking any aspec o he conce n.
7. Di e ence be ween RCA ools and echniques and RCA me hods
To di e en ia e Roo Cause Analysis Tools and Roo Cause Analysis Me hods, a
s anda d is needed o which hey could be compa ed. I is gene ally ag eed ha
he pu pose o oo cause analysis is o ind e ec i e solu ions o ou p oblems
such ha hey do no ecu . Acco dingly, an e ec i e oo cause analysis p ocess
should p o ide a clea unde s anding o exac ly how he p oposed solu ions mee
his goal.
To p o ide his assu ance an e ec i e p ocess should mee he ollowing six c i e ia
 Clea ly de ines he p oblem and i s signi icance o he p oblem owne s.
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 33
D. Mah o; A. Kuma
 Clea ly delinea es he known casual ela ionships ha combined o cause
he p oblem.
 Clea ly es ablishes causal ela ionships be ween he oo causes and he
de ined p oblem
 Clea ly p esen s he e idence used o suppo he exis ence o iden i ied
causes.
 Clea ly explains how he solu ions will p e en ecu ence o he de ined
p oblem.
 Clea ly documen s c i e ia 1 h ough 5 in inal RCA epo so o he s can
easily ollow he logic o he analysis
The e o e, he e is a clea dis inc ion be ween an RCA Tool and RCA Me hod. A ool
is dis inguished by i s limi ed use pe aining o pa icula phenomena o si ua ion,
while a me hod may in ol e many s eps and p ocesses and has wide usage wi h
he lexibili y o modi y o some ex en pe aining o pa icula phenomena o
si ua ion. Compa a i e di e ences o selec ed RCA Tools and RCA Me hods ha e
been shown in he Table 1.
Tool / Me hod Type De ines
P oblem
De ines
all causal
ela ion-
ships
P o ides
a causal
pa h o
oo
causes
Delinea es
e idence
Explains
how
solu ions
p e en
ecu ence
Easy o
ollow
epo
Causes-and-E ec
Diag am Tool Yes Limi ed No No No No
In e ela ionship
Diag am Tool Yes No No No No No
Cu en Reali y
T ee Tool Yes No Limi ed No Limi ed No
Why Why Analysis Tool Yes No Yes No No No
Mul i Va i Analysis Tool Limi ed Limi ed Yes No No Yes
E en s and Causal
Fac o Analysis Me hod Yes Limi ed No No No No
Change Analysis Me hod Yes No No No No No
Ba ie Analysis Me hod Yes No No No No No
Managemen
O e sigh and Risk
T ee Analysis Me hod Yes Yes Yes No Limi ed Yes
Human
Pe o mance
E alua ion Me hod Yes Yes Yes No Limi ed Yes
Kepne – T egoe
P oblem Sol ing
and Decision
Making
Me hod Yes Yes Yes No Limi ed Yes
Table 1. “Compa ison o selec ed RCA ools and RCA me hods”.
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 34
D. Mah o; A. Kuma
8. Roo cause analysis (RCA) p ocess
The RCA me hod b ings a eam o , usually 3 o 6 o as demanded, knowledgeable
people oge he o in es iga e he ailu e using e idence le behind om he aul .
The eam b ains o ms o ind as many causes o he aul as possible. By using
wha e idence emained a e he aul and h ough discussions wi h people
in ol ed in he inciden , all he non-con ibu ing causes a e emo ed and he
con ibu ing causes e ained.
A aul ee is cons uc ed s a ing wi h he inal ailu e and p og essi ely acing
each cause ha led o he p e ious cause. This con inues ill he ail can be aced
back no u he . Each esul o a cause mus clea ly low om i s p edecesso ( he
one be o e i ). I i is clea ha a s ep is missing be ween causes i is added in and
e idence looked o o suppo i s p esence. Once he aul ee is comple ed and
checked o logical low, he eam hen de e mines wha changes has o be made
o p e en he sequence o causes and consequences om again occu ing.
Roo cause analysis is de ined in he Canadian Roo Cause Analysis F amewo k1 as
“an analy ic ool ha can be used o pe o m a comp ehensi e, sys em-based
e iew o c i ical inciden s. I includes he iden i ica ion o he oo and con ibu o y
ac o s, de e mina ion o isk educ ion s a egies, and de elopmen o ac ion plans
along wi h measu emen s a egies o e alua e he e ec i eness o he plans.”
Roo cause analysis in indus ies is bes conduc ed by a mul idisciplina y eam,
in ol ing indi iduals knowledgeable abou P oduc i i y, as well as knowledgeable in
he Quali y a ea o ocus. In o ma ion is ga he ed h ough in e iews wi h s a
membe s who we e di ec ly and indi ec ly in ol ed, as well as amily membe s
when possible. In addi ion, he eam e iews he loca ion whe e he inciden
occu ed, examines he p oduc s, de ices, en i onmen and wo k p ocesses
in ol ed, and e iews ele an documen a ion and li e a u e.
To imp o e p oduc quali y and p oduc i i y, he analysis eam p oceeds h ough a
se ies o p obing ques ions ocused on answe ing “why” and “caused by” ques ions
o delinea e he a ious ac o s ha con ibu ed o he e en and which, i le
unmi iga ed, could con ibu e o ano he e en . The ocus is on sys ems and
p ocesses and hei in e ac ion wi h indi iduals, wi h he unde s anding ha he
indi iduals in ol ed did no in en ionally ac o cause ha m, and gi en he same se
o ci cums ances, he ou come would be he same o any indi iduals in ol ed. The
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 35
D. Mah o; A. Kuma
oo cause analysis p ocess encou ages high-le e age sys em changes ha , i
implemen ed, will ha e las ing e ec s on p oduc quali y and p oduc i i y wi h
sa e y.
Rele an li e a u e and p ac ice s anda ds a e conside ed in o mula ing
ecommenda ions and ac ions. To make i unde s and he signi icance o such a
sys em enhancemen , i has been p o ided an analogous example om he
au omo i e indus y conside ing he s eps o success ul oo cause analysis as
gi en below.
8.1 S eps o success ul oo cause analysis (RCA)
Roo Cause Analysis (RCA) is a use ul ool o ouble shoo ing b eakdowns and
e icien ly coming o a solu ion. Fo success ul implemen a ion o RCA ollowing
se en s eps a e necessa y and once comple ed ha will na u ally esul in
elimina ion o oo causes and will inc ease p o i s.
Se en poin s o RCA a e…
 Desc ibe he ac ual Cause.
 De ine he physical phenomena o he Cause
 O ganize he de ails o he Cause by using he '3W2H' (wi h wha , when,
whe e, how, how much) ool.
 Wo k as a eam, espec ing each o he ’s expe ise and knowledge, a he
han indi idually.
 Conside e e y possible cause o he Cause.
 Ve i y all logical causes and elimina e all illogical causes.
 I de e mined ha he cause among he causes was human e o , sepa a e
ha cause om he physical causes.
9. Ou line o he empi ical case s udy
The expe imen a ion was ca ied ou a one o he au omo i e componen
manu ac u ing plan s si ua ed in Jamshedpu , India. The plan has sophis ica ed

doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 36
D. Mah o; A. Kuma
mode n machine ools. This plan is p o essionally managed and i has go ISO
9002 and ISO 14001-sys em ce i ica ion. I has also implemen ed To al Quali y
Managemen . The depa men s ha e been compu e ized and linked h ough Local
A ea Ne wo king (LAN) o enhance accessibili y. Main enance managemen
sys ems a e in p ac ice. I is also backed up by an ad anced compu e aided
condi ion moni o ing sys em. The plan p ocesses may be oughly ca ego ized in o
he ollowing blocks (as pe he ma e ial low):
 Raw Ma e ial P ocesses
 In e media e P ocesses
 Final P ocesses
The p oduc s manu ac u ed by he plan comp ise o di e en componen s, used in
cons uc ion equipmen s and in con eying sys ems. To imp o e he p oduc quali y
and p oduc i i y Oxy Flame cu ing machine was chosen. This machine is
composed o elec ical and mechanical sys ems. The machine mo es along X axis
h ough L.T. Mechanism (Long T a el Mechanism) and h ough c oss a el i.e.
along Y axis a el is done in c oss a el Beam by he h ee cu ing o ches which
a e i ed wi h senso s o which command comes om he CPU i ed wi h he
machine. This machine has h ee cu ing o ches by which he ma e ials a e being
cu by oxy lame. I uses mul i channel da a o ma o s o age o ime da a,
spec a, e c., including: unc ion iden i ie , sampling equency, inpu /ou pu poin
and di ec ion, inpu /ou pu uni s, ee ex lines, X-, Y-, Z-axis labels, auxilia y
cus om ields. Ex ensi e commands o ex ac in o ma ion om heade s including a
sea ch unc ion. I uses dissol ed Ace ylene and Oxygen o gene a e lame o cu
he ma e ial in a s aigh line o in cu es. When a single o ch is used, i can cu
s aigh be eling. The o ches ha e p oximi y senso s so ha he e should be a
accu a e dis ance be ween he Raw ma e ial o be cu and To ch ip. The ool
holde is ha pa o he sys em whe e he senso o he cu ing edge is placed.
When he machine begins cu ing, he cu ing o ches mo e acco ding o DNC
(Di ec Nume ical Con ol) p og amme. This DNC p og amme is con e ed om a
CNC (Compu e ized Nume ical Con ol) p og amme o cu ing. A de ec o is also
placed o cap u e he p oblems in cu ing, which enables an ope a o o make
adjus men s o speed, cu ing gap o Gas low. A pho og aph o he machine is
p esen ed in Figu e 1.
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 37
D. Mah o; A. Kuma
Now he se en poin s as desc ibed abo e has been adop ed o s udy he empi ical
case.
9.1 Finding and De ining he Ac ual Cause (S ep 01)
The p oblem was encoun e ed in he ini ial p ocessing o he ma e ial. The capaci y
u iliza ion o he plan was a ound 55% o 65% due o p oblems in p ocessing o
ma e ials i sel and he e was always i e- igh ing o wan o ma e ial. Bu , no oo
causes we e iden i ied as o why he e was such a p oblem. This eason was one o
he key con ibu o y ac o s o he lowe le el o p oduc i i y
Figu e 1. “Sample machine aken o oo cause analysis”.
9.2 Physical phenomena o he cause (S ep 02)
The cu ing ope a ion was o be pe o med in all he i ems bu how a i is ela ed
o o i was in luencing he p oduc ion p ocesses had ne e been s udied ea lie .
The e o e, a alue s eam mapping was done i s by selec ing a job (Pi o F ame)
o unde s and he pe cen ile impac o gas cu ing ope a ion on he p oduc ion
p ocess. The de ec on cu ma e ial inc eases he cycle ime o each ac i i y and
adds mo e non- alue adding imes. Hence, o pin poin non- alue adding ac i i ies
con ibu ed by gas cu ing, da a we e cap u ed ac i i y wise.
Now, he Value S eam Mapping was done and i s de ail is abula ed in Table 2.
Di e en ope a ions ha e been ca ego ized and en e ed in he able in abb e ia ed
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 38
D. Mah o; A. Kuma
e ms. The de ails o he abb e ia ion used a e gi en below. The same
abb e ia ions shall be used he e in a e .
 VA = Value Adding Ac i i ies
 NVA = Non Value Adding Ac i i ies
 EOT = Elec ic Ope a ed Towe C ane
 DNC = Di ec Nume ical Con ol sys em
 CNC = Compu e ized Nume ical Con ol sys em
 DTD = Desk o Desk (Lean ool)
 SPC = S a is ical P ocess Con ol
Value Adding
Non Value Adding
Ma e ial Incoming
S o e
Handling / Se ing
Gas Cu ing
Cu ing Pieces S o age
G inding / Cleaning
Handling / Se ing
Bending
Handling / Se ing
Assembly
Handling / Se ing
Welding
Cleaning
Tu ning
Handling / Se ing
Bo ing
Handling / Se ing
D illing
Handling / Se ing
Cleaning / Oiling
Dispa ch
1.5
0.3
2 4 2
3 1.5
2 1
5
0.5
5
2
4
0.3
4
0.3
4 0.3
1 2
SUMMARY
To al
Time (H )
46
Value
Adding (H ) 30
Non Value
Adding (H ) 16
No e: In p ocess Inspec ion is ca ied ou a e e y s age
Figu e 2. “Value s eam mapping be o e oo cause analysis”.
RAW
MATERIAL
FINISHED
PRODUCT
TIME H
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 39
D. Mah o; A. Kuma
In he abo e sample s udy, i was ound ha he non- alue adding ac i i ies we e
highe han he alue adding ac i i ies. The ope a ions en e ed in se ial numbe s 3
o 5, 7 o 12, 18 o 20, 25 and 26 (Table 2) a e non- alue adding ac i i ies
associa ed o gas cu ing ope a ion. I comes ou o be 07 hou s 40 minu es o
non- alue added ac i i y wi h o al h oughpu ime o 12 Hou s 50 minu es.
Sl
No Ope a ions in
Sequence Machine wise Ac i i y
Desc ip ion Resou ces
in ol ed
Time
aken
(Min) Ca ego y
1 Raw Ma e ial
handling
Handling om s ock ya d
o Co ina Machine
EOT C ane, 1
Ope a o , 1 Helpe 25 NVA
2 Da a con e sion DNC o CNC Co ina
Machine
1 Enginee , 1
Compu e 10 NVA
3 CNC Cu ing A Co ina Machine 1 Ope a o 35 VA
4
Ma e ial Remo al
shi ing and
Inspec ion
Co ina Machine 2 helpe , 1 C ane, 1
Inspec o 55 NVA
5 Ma e ial P epa a ion Manual g inding 2 Ope a o , 2
G inding machine 25 NVA
6 Inspec ion Manual g inding 1 Inspec o 20 NVA
7 Seg ega ion &
Shi ing
Ma e ial P epa a ion,
Bending, Assembly o
Machining
1 Ope a o , 1
Helpe , o k li s,
olleys, C ane
30 NVA
8 Assembly
Collec ion o p epa ed
ma e ial o assembly
1 Ope a o , 2
Helpe , Fix u es,
Gauges
40 VA
9 Inspec ion Assembly 1 Inspec o 20 NVA
10 Loading & se ing a
Manipula o Fo welding EOT C ane, 1
Ope a o 15 NVA
11 Welding Mig welding 1 Ope a o , Co2
Gas, Welding M/c 60 VA
12 Inspec ion Manually, UT machine 1 Inspec o 25 NVA
13 Unloading & Shi ing
o Machining cen e Unloading by EOT C ane 1 Helpe , Fo k li 20 NVA
14 Se ing a VTL Fo machining 1 Helpe , 1 EOT
C ane, 1 Ope a o , 20 NVA
15 Base Machining Fix u e and special ool 1 Ope a o 50 VA
16 Inspec ion Ve nie , Jig 1 Inspec o 20 NVA
17 Unloading & Shi ing
o Bo ing Fo k li , EOT C ane 1 Helpe , Fo k li ,
EOT C ane 20 NVA
18 Se ing a Ho izon al
Bo ing Fo Machining 1 Helpe , EOT
C ane, 1 Ope a o , 20 NVA
19 Bo ing Ø 90±1., Ø 80±1 1 Ope a o 50 VA
20 Inspec ion Ve nie , Jig 1 Inspec o 20 NVA
21 Unloading & Shi ing
o D illing & Tapping Fo k li , EOT C ane 1 Helpe , 20 NVA
22 Se ing a Radial D ill D illing Fixing o Jig 1 Helpe , EOT
C ane, 1 Ope a o 20 NVA
22 D illing & Tapping Ø 15.5+0.2, Ø 20+0.3 1 Ope a o 75 VA
23 Inspec ion Gauge, Tap 1 Inspec o 25 NVA
24 Unloading & Shi ing
o cleaning Fo k li , EOT C ane 1 Helpe , Fo k li ,
EOT C ane 15 NVA
25 Cleaning & su ace
ea men Phospha ing and Rus oil 1 Ope a o , 1
helpe 25 NVA
26 Inspec ion Visually 1 Inspec o 10 NVA
To al ∑770
Table 2. “Value s eam mapping be o e oo cause analysis”.
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 46
D. Mah o; A. Kuma
Sl No Ope a ion in
Sequence Machine wise
Ac i i y Desc ip ion Resou ces
in ol ed
Time
aken
(Min)
Ca ego y
(VA/
NVA)
1 Raw Ma e ial
handling
Handling om s ock
ya d o Co ina M/c
1 Helpe , EOT
C ane, 1 Ope a o 15 NVA
2 Da a con e sion DNC o CNC Co ina
Machine
1 Enginee , 1
Compu e 10 NVA
3 CNC Cu ing A Co ina Machine 1 Ope a o 30 VA
4 Ma e ial Remo al &
shi ing F om Co ina Machine 2 helpe , 1 C ane 10 NVA
5 Inspec ion A co ina machine 1 Inspec o 15 NVA
6 Assembly
Ma e ial p epa ed a
di e en loca ion a e
ga he ed & assembled
1 Ope a o , 2
Helpe , Fix u es,
Gauges
40 VA
7 Loading & se ing a
Manipula o Fo welding EOT C ane, 1
Ope a o 7 NVA
8 Welding Mig welding
1 Ope a o , Co2
Gas, Welding
Machine
50 VA
9 Inspec ion UT machine 1 Inspec o 15 NVA
10
Unloading &
Shi ing o
Machining
Fo k li , EOT C ane 1 Helpe , Fo k li 20 NVA
11 Se ing a VTL Fo machining 1 Helpe , EOT
C ane, 1 Ope a o 20 NVA
12 Base Machining Fix u e & special ool 1 Ope a o 45 VA
13 Inspec ion Ve nie , Jig 1 Inspec o 20 NVA
14 Unloading &
Shi ing o Bo ing Fo k li , EOT C ane 1 Helpe , Fo k li ,
EOT C ane 20 NVA
15 Se ing a
Ho izon al Bo ing Fo machining 1 Helpe , EOT
C ane, 1 Ope a o 20 NVA
16 Bo ing Ø 90±1., Ø 80±1 1 Ope a o 30 VA
17
Unloading &
Shi ing o D illing
& Tapping
Fo k li , EOT C ane 1 Helpe , Fo k li ,
EOT C ane 15 NVA
18 Se ing a Radial
D ill Fo D illing 1 Helpe , EOT
C ane, 1 Ope a o 20 NVA
19 D illing & Tapping Ø 15.5+0.2, Ø 20+0.3 1 Ope a o 60 VA
20 Inspec ion Gauge, Tap 1 Inspec o 25 NVA
21 Unloading &
Shi ing o cleaning Fo k li , EOT C ane 1 Helpe , Fo k li ,
EOT C ane 10 NVA
22 Cleaning & su ace
ea men Phospha ing & Rus oil 1 Ope a o , 1
helpe 25 NVA
To al 507
Table 8. “Value s eam mapping a e oo cause analysis”

doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 47
D. Mah o; A. Kuma
Figu e 5. “T end analysis o cu ing p oduc s a e oo cause elimina ion”.
The alue s eam mapping was done a e oo cause elimina ion, which has been
shown in Table 8 and Figu e 7 espec i ely.
Figu e 6. “Compa a i e end analysis o cu ing p oduc s be o e a e Roo Cause analysis”.
155 178 165 170 188 190 196 191 198 195 200 198 201
145 170 160 167 184 188 194 189 196 195 197 197 201
10 853422220310
0
50
100
150
200
250
12345678910111213
No o i ems
Days :Janua y 2008
T end analysis o P oduc Quali y a e Roo cause analysis
To al I em Cu Accep ed Rejec ed
6.90 4.71 3.13 1.80 2.17 1.06 1.03 1.06 1.02 0.00 1.52 0.51 0.00
8.28
6.47
11.25 9.58
11.96 12.23
9.79
12.70 12.24
15.90 16.24
13.71 13.93
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
18.00
12345678910111213
Pe cen age
Sample da a aken o days
Compa a i e end analysis o p oduc quali y
A e S udy
Be o e S udy
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 48
D. Mah o; A. Kuma
New S a e P ecedence Diag am
Value Adding Non Value Adding
Incoming
g
Se ing
Gas Cu ing
g
Cleaning
g
Se ing
Bending
g
Se ing
Assembly
g
Se ing
Welding
g
G inding
Tu ning
g
Se ing
Bo ing
Se ing
D illing
g
Oiling
Dispa ch
1.5 2 2.5 3 1.5 2 1 2.5
0.5
3.5
2 3 0.2
4 0.2 4 0.3
2
SUMMARY
To al
Time (H ) 36 Value
Adding (H ) 24
Non Value
Adding (H ) 12
No e: In p ocess Inspec ion is ca ied ou a e e y s age
Figu e 7. “Value S eam Mapping a e oo cause analysis o dump le e ”.
Table 9 ep esen s he compa ison o esul s o pi o ame be o e and a e he
oo cause analysis.
Sl No S a us Be o e oo cause A e oo cause
elimina ion % Change
1 Time in
Minu es To al Time % To al Time % (+ O -)
2 Value
Adding 310 40.25 260 51.28 +16.13
3 Non Value
Adding 460 59.75 247 48.72 -46.30
Table 9. “Compa ison o esul s o pi o ame”
12. Compa ison o o ch quali ies and p oduc i i y
A e implemen a ion o oo cause analysis and elimina ion o de ec s, i has been
obse ed ha he quali y end has imp o ed a lo , bu s ill some p oblems ha e
been le , which needs o be add essed.
TIME
H
RAW
MATERIAL FINISHED
PRODUCT
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 49
D. Mah o; A. Kuma
In o de o es he a ia ion in quali y and p oduc i i y o Oxy Flame Cu ing
Machine by h ee o ches A, B and C, da a has been ga he ed o all he shi s.
The ea e , Mul i Va i Analysis app oach was adop ed. I is impe a i e he e o
men ion ha Mul i Va i Analysis ies o ind he ela ion among cyclic e ec s,
empo al e ec s and posi ional e ec .
An ANOVA es has been ca ied ou o es he p oduc i i y con ibu ion o each
o ch s a is ically. Howe e , his ANOVA es may also be ca ied ou by MINITAB 15
so wa e.
The da a has been collec ed o Janua y 2008. The weekly a e ages o no o good
pieces p oduced by all he cu ing o ches ha e been wo ked ou , which has been
abula ed in Table 10.
P oduc ion a e ages Week 1 Week 2 Week 3 Week 4 Week 5
To ch A 220 251 226 246 260
To ch B 244 235 232 242 225
To ch C 252 272 250 238 256
Table 10. “Weekly a e ages o p oduc ion by he o ches”.
We ha e m independen samples, whose size is n and whe e he membe s o he
i h sample – Xi1, Xi2, …, Xin a e no mal andom a iables wi h unknown mean µi
and unknown a iance 2

, as he Equa ion 1 shows:
Xij ~N(

i,

2), whe e, i=1…m and j=…n
Equa ion 1. “Samples”.
The hypo hesis o be es ed is he p oduc i i y con ibu ion is he same in he h ee
o ches o no . Le ,
 H0: µ1 = µ2 = …………=µm
 H1≠ equal /i.e. means a e no equal.)
The algeb aic iden i y, called he sum o squa es iden i y, is use ul in doing hese
ypes o compu a ions. The Equa ion 2 shows he s a is ic o compu ing sum o
squa e iden i y:
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 50
D. Mah o; A. Kuma
wb
m
i
n
jij SSSSnmXX 


22
11
)(
Equa ion 2. “S a is ic o compu ing sum o squa e iden i y”.
Whe e SSb and SSw a e calcula ed as he Equa ion 3 and Equa ion 4 shows:
SSw(Xij
j1
n

i1
m
Xi)2
Equa ion 3. “Wi hin samples sum o squa es”.
SSbn(Xi
i

1
m
X2
)
Equa ion 4. “Be ween samples sum o squa es”.
The abo e able can be simpli ied by sub ac ing each alue wi h 220 ha
sub ac ing a cons an om each da a; alue will no a ec he alue o es
s a is ic. Hence he new Table 10 can be o med as ollows
P oduc ion
a e ages Week 1 Week 2 Week 3 Week 4 Week 5

jij
X jij
X2
To ch A 031 626 40 103 3273
To ch B 24 15 12 22 578 1454
To ch C 32 52 30 18 36 168 6248
Table 11. “Modi ied weekly a e ages o p oduc ion by he o ches”.
F om he Table 11 (whe e m=3 and n=5) and he Equa ions 1-4, we ob ain a alue
o 2.60 in he es s a is ic, as Equa ion 5 shows:
TS = 60.2
12
5785.19912
3335.863
1


mnm
ss
m
ss
w
b
Equa ion 4. “The alue o es s a is ic (m=3 and n=5)”.
12.1 In e ence om S a is ical Calcula ion
Now, by compa ing he calcula ed alue wi h abula ed alue o F.05,n,m, we see
ha F2.12,.05 = 3.89. Hence, because he alue o he es s a ic does no exceed
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 51
D. Mah o; A. Kuma
3.89, i canno be, a 5 pe cen le el o signi icance, ejec ed he null hypo hesis
ha he o ches gi e equal p oduc ion i.e. all he o ches p oduce equal quali y and
p oduc i i y a 5 pe cen le el o signi icance.
13. Conclusions
The con en ional Roo Cause Analysis Tools and Me hods p o ide some s uc u e o
he p ocess o human e en p oblem sol ing. This empi ical s udy shows as o how
hey can be used and how i can be communica ed o o he s wi h ull app ecia ion.
How he solu ions will p e en he p oblem om ecu ing. Thus, i is he only
p ocess which allows all s akeholde s o ha e a clea idea and he eali y o
p omo e i s e ec i e solu ion all he ime. The Roo Cause Tools and Me hods could
be u ilized acco ding o p e alen condi ions and si ua ions o Man, Ma e ial,
Machines, Sys ems and P ocesses.
In he amewo k o his s udy, he ollowing conclusions can be d awn
 I has been obse ed ha he pe cen age inc ease in alue adding is +
20.00 %, whe eas, pe cen age educ ion in non- alue adding is –25.00 %
a e implemen a ion o oo cause analysis o dump le e .
 I has been obse ed ha he pe cen age inc ease in alue adding is +
16.13 %, whe eas, pe cen age educ ion in non- alue adding is –46.30 %
a e implemen a ion o oo cause analysis o pi o ame.
 The ejec ion has educed om 11.87% o 1.92% on an a e age due o
a aining he skill wi hin a e y sho ime a e implemen a ion o p ope
main enance schedule and gi ing aining o he ope a o s and main enance
pe sons.
 I has been obse ed ha a e applica ion o Roo Cause Analysis, he
p oduc quali y and p oduc i i y o he plan has imp o ed. The plan has
now ixed a e y high a ge , om 250 T pe mon h o 450 T pe mon h,
which may no be possible in he p e ailing si ua ion. The company will
ha e o achie e he capaci y u iliza ion o un ac o o 77% ins ead o
ea lie s a ed alue o 55 o 65 % o i .

doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 52
D. Mah o; A. Kuma
 All he o ches A, B and C p oduce same le el o quali y and p oduc i i y a
5% le el o signi icance.
Re e ences
Ande son, B., & Fage haug, T., (2000). Roo Cause Analysis: Simpli ied Tools and
Techniques. Milwaukee: ASQ Quali y P ess.
A ca o, J. S. (1997). TQM Facili a o ’s Guide. Boca Ra on,FL: S . Lucie P ess.
B assa d, M., & Ri e , D. (1994).The Memo y Jogge II: A Pocke Guide o Tools
o Con inuous Imp o emen and E ec i e Planning. Salem, NH: GOAL/QPC.
B assa d, M. (1996). The Memo y Jogge Plus+: Fea u ing The Se en Managemen
and Planning Tools. Salem, NH: Goal/QPC.
B own, J. I. (1994). Roo -Cause Analysis in Comme cial Ai c a Final Assembly.
Mas e ’s Abs ac s In e na ional, 33(6), 1901-.
Cox, J. F. III, & Spence , M. S. (1998). The Cons ain s Managemen Handbook.
Boca Ra on, FL: S .Lucie P ess.
Dean, L. G. (2007). Compa ison o Common Roo Cause Analysis Tools and
Me hods. Apollo Roo Cause Analysis – A new way o Thinking, 3 d Edi ion.
De me , H. W. (1997). Gold a ’s Theo y o Cons ain s. Milwaukee: ASQC P ess.
Dew, J. R. (1991). In Sea ch o he Roo Cause. Quali y P og ess, 24(3), 97-107.
DOE Guideline (1992). Roo Cause Analysis Guidance Documen . US Depa men o
Ene gy: Washing on.
Dugge , A. M. (2004). A s a is ical Compa ison o Th ee Roo Cause Tools. Jou nal
o Indus ial Technology, 20(2), 1-9.
F edendall, L. D., Pa e son, J. W., Lenha z, C., & Mi chell, B. C. (2002). Wha
Should Be Changed?. Quali y P og ess, 35(1), 50-59.
Ins i u e o Sa e Medica ion P ac ice (2005). Fluo acil Inciden Roo Cause Analysis:
Final D a .
doi:10.3926/jiem.2008. 1n2.p16-53 ©© JIEM, 2008 – 01(02):16-53 - ISSN: 2013-0953
Applica ion o oo cause analysis in imp o emen o p oduc quali y and p oduc i i y 53
D. Mah o; A. Kuma
Lepo e, D., Cohen, O., Deming & Gold a E. (1999). The Theo y o Cons ain s and
he Sys em o P o ound Knowledge. G ea Ba ing on, MA: No h Ri e P ess.
Mo an, J. W., Talbo , R. P., & Benson, R. M. (1990). A Guide o G aphical P oblem-
Sol ing P ocesses. Milwaukee: ASQC Quali y P ess.
Pasqua ella, M., Mi chell, B., & Sue ken, K. (1997). A Compa ison on Thinking
P ocesses and To al Quali y Managemen Tools. In 1997 APICS Cons ain s
Managemen P oceedings: Make Common Sense: A Common P ac ice. Den e ,
CO: Falls Chu ch, VA: APICS, 59-65.
Robson, M. (1993). P oblem Sol ing In G oups. (2nd ed.). B ook ield, VT: Gowe .
Scheinkop , L. J. (1999). Thinking Fo A Change: Pu ing The TOC Thinking
P ocesses o Use. Boca Ra on, FL:S . Lucie P ess.
Smi h, D. (2000). The Measu emen Nigh ma e: How he Theo y o Cons ain s Can
Resol e Con lic ing S a egies, Policies, and Measu es. Boca Ra on, FL: S . Lucie
P ess.
Sp oull, B. (2001). P ocess P oblem Sol ing: A Guide o Main enance and
Ope a ions Teams. Po land: P oduc i i y P ess.
Wilson, P. F., Dell, L. D., & Ande son, G. F. (1993). Roo Cause Analysis: A Tool o
To al Quali y Managemen . Milwaukee: ASQC Quali y P ess.
©© Jou nal o Indus ial Enginee ing and Managemen , 2008 (www.jiem.o g)
A icle's con en s a e p o ided on a A ibu ion-Non Comme cial 3.0 C ea i e commons license. Reade s a e
allowed o copy, dis ibu e and communica e a icle's con en s, p o ided he au ho 's and Jou nal o Indus ial
Enginee ing and Managemen 's names a e included. I mus no be used o comme cial pu poses. To see he comple e
license con en s, please isi h p://c ea i ecommons.o g/licenses/by-nc/3.0/.