Professional Documents
Culture Documents
n
i = 1
n
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 49
FO RECASTI N G Ti me Ser i es
The f ol l ow i ng data show t he number of l i t er s of gasol i ne sol d by Pet r on i n Bagui o Ci t y f or t he past 12 w eek s.
W eek
Act ual
D emand*
For ecast *
Nave SM A W M A ES, = 0.30 Li near t r end
F | A - F| F | A - F| F | A - F| F | A - F| F | A - F|
1 17
2 21
3 19
4 23
5 18
6 16
7 20
8 18
9 22
10 20
11 15
12 22
13
M AD
* 1,000 l i t er s
1. For ecast demand usi ng t he nave appr oach.
2. For ecast demand usi ng 3 - w eek si mpl e movi ng aver age model .
3. Usi ng a w ei ght of 0.40 f or t he most r ecent obser vat i on, 0.30 f or t he second most r ecent , 0.20 f or t he t hi r d most
r ecent , and 0.10 f or t he f our t h most r ecent , f or ecast demand usi ng 4- w eek w ei ght ed movi ng aver age model .
4. For ecast demand usi ng ex ponent i al smoot hi ng model w i t h = 0.30.
5. For ecast demand usi ng l i near t r end anal ysi s.
6. Cal cul at e t he M AD f or each of t he above f or ecast i ng model s. W hi ch f or ecast i ng model w oul d you r ecommend
t o Pet r on s manager ? W hy?
7. W hat demand f or ecast w oul d you r ecommend f or w eek 13?
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 50
FO RECASTI N G Li near Regr essi on
D on Henr i co s Pi zza i s i nt er est ed i n est abl i shi ng t he r el at i onshi p bet w een i t s adver t i si ng and sal es on t he
f ol l ow i ng dat a:
Q uar ter
Adver t i si ng
( P100,0 00)
Sal es
( PM i l l i on)
1 4 1
2 10 4
3 15 5
4 12 4
5 8 3
6 16 4
7 5 2
8 7 1
9 9 4
10 10 2
1. Can Don Henr i co s Pi zza use l i near r egr essi on anal ysi s t o f or ecast i t s f ut ur e sal es gi ven a pl anned adver t i si ng
ex pendi t ur e?
2. D et er mi ne t he est i mat ed equat i on of t he r egr essi on l i ne t hat def i nes t he r el at i onshi p bet w een sal es and
adver t i si ng.
3. I f D on Henr i co s Pi zza adver t i si ng ex pense i s ex pect ed t o be P1,100,000.00, w hat w oul d be t he cor r espondi ng
sal es f or ecast?
4. I f D on Henr i co s Pi zza spends f i ve per cent of sal es on pi zza box es, how much box es, i n peso vol ume, shoul d i t s
oper at i ons manager or der f or t he nex t quar t er ?
Q uanti tati ve For ecasti ng
The f ol l ow i ng t abl e r epr esent s sal es dat a f or pal m oi l ( i n hundr ed gal l ons) sol d by a gr ocer y st or e.
M onth Sal es 1. Gi ven a choi ce on w het her t o use t he nave appr oach; a 5- mont h SM A
model ; a W M A model w i t h w ei ght s 0.20, 0.30, and 0.50; and an ES
model w i t h = 0.60 and a f or ecast of 3,600 gal l oons i n t he f i r st
mont h; w hi ch model w oul d you use? W hy?
2. Based on your deci si on i n i t em 1, w hat shoul d be t he f or ecast f or
mont h 11?
3. W oul d you r ecommend usi ng l i near r egr essi on f or ecast i ng model f or
t hi s t i me ser i es dat a? W hy?
4. I f t he l i near r egr essi on may be used, w hat w oul d be t he f or ecast f or
mont h 16?
1
2
3
4
5
6
7
8
9
10
40
48
40
45
50
49
46
57
54
62
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 51
E. TOTAL QUALITY MANAGEMENT
Lear ni ng Obj ect i ves Af ter compl et i ng th i s topi c, you shoul d be abl e to:
1. D escr i be TQM .
2. Gi ve an over vi ew of pr obl em sol vi ng.
3. Gi ve an over vi ew of pr ocess i mpr ovement .
4. D escr i be and use var i ous qual i t y t ool s.
Total Q ual i ty M anagement a phi l osophy t hat i nvol ves ever yone i n an
or gani zat i on i n a cont i nual ef f or t t o i m pr ove qual i t y and achi eve cust omer
sat i sf act i on.
2 Key Phi l osophi es
1. Cont i nuous i mpr ovement seek s t o mak e never - endi ng
i mpr ovement s t o t he pr ocess of conver t i ng i npu t s i n t o
out put s.
2. Customer sati sf acti on i nvol ves meet i ng or ex ceedi ng
cust omer ex pectat i ons.
D escr i pt i on of t he TQM Appr oach
1. Fi nd ou t w hat customer s w ant. I nt egr at e t he cust omer s voi ce i n t he deci si on- mak i ng pr ocess. Be sur e t o
i ncl ude t he i n t er nal cust omer ( t he nex t per son i n t he pr ocess) as w el l as t he ex t er nal cust omer ( t he f i nal
cust omer ) .
2. D esi gn a pr oduct or ser vi ce that w i l l meet or ex ceed w hat customer s w ant. M ak e i t easy to use and easy t o
pr oduce.
3. D esi gn a pr oducti on pr ocess that f aci l i tates doi ng t he j ob r i ght the f i r st the f i r st ti me. D et er mi ne w her e
mi st ak es ar e l i k el y t o occur and t r y t o pr event t hem. W hen mi st ak es do occur , f i nd out w hy so t hat t hey ar e l ess
l i k el y t o occur agai n. St r i ve t o mak e t he pr ocess mi stak e- pr oof .
4. Keep tr ack of r esul ts, and use those to gui de i mpr ovemen t i n the system. Never stop t r yi ng t o i mpr ove.
5. Ex tend these concepts to suppl i er s and to di str i but i on.
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 52
Ot her El ement s of t he TQM Appr oach
1. Cont i nual i mpr ovement
2. Compet i t i ve benchmar k i ng
3. Empl oyee empow er ment
4. Team appr oach
5. D eci si ons based on f act s r at her t han opi ni ons
6. Know l edge of qual i t y tool s
7. Suppl i er qual i t y
Qual i t y at t he sour ce
Compar i ng t he Cul t ur es of TQM vs. Tr adi t i onal Or gani zat i ons
ASPECT TRAD I TI ONAL TQM
Over al l mi ssi on
Obj ect i ves
M anagement
Rol e of manager
Cust omer r equi r ement s
Pr obl ems
Pr obl em sol vi ng
I mpr ovement
Suppl i er s
Jobs
Focus
M ax i mi ze r et ur n on i nvest ment
Emphasi s on shor t t er m
Not al w ays open; somet i mes
i nconsi st ent obj ect i ves
I ssue or der s; enf or ce
Not hi ghest pr i or i t y; maybe
uncl ear
Assi gn bl ame; puni sh
Not syst emat i c; i ndi vi dual s
Er r at i c
Adver sar i al
Nar r ow , speci al i zed; much
i ndi vi dual ef f or t
Pr oduct or i ent ed
M eet or ex ceed cust omer
sat i sf act i on
Bal ance of l ong t er m and shor t
t er m
Open; encour ages empl oyee
i npu t ; consi st ent obj ect i ves
Coach, r emove bar r i er s, bui l d
t r ust
Hi ghest pr i or i t y; i mpor t ant t o
i dent i f y and under st and
I dent i f y and r esol ve
Syst emat i c; t eams
Cont i nual
Par t ner s
Br oad, mor e gener al ; much t eam
ef f or t
Pr ocess or i ent ed
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 53
Pr obl em Sol vi ng
To be successf ul , pr obl em- sol vi ng ef f or t s shoul d f ol l ow a st andar d
appr oach.
An i mpor t ant aspect of pr obl em sol vi ng i n t he TQM appr oach i s
el i mi nati ng the cause so t hat t he pr obl em does not r eoccur .
User s of t he TQM appr oach of t en l i k e t o t hi n k of pr obl ems as
oppor t uni t i es f or i mpr ovement .
Basi c St eps i n Pr obl em Sol vi ng
Step 1 D ef i ne the pr obl em and establ i sh an i mpr ovement goal .
Gi ve pr obl em def i n i t i on car ef ul consi der at i on; don t r ush t hr ough t hi s st ep because t h i s w i l l ser ve as
t he f ocal poi nt of pr obl em- sol vi ng ef f or t s.
Step 2 Col l ect data.
The sol ut i on must be based on f acts. Possi bl e t ool s i ncl ude check sheet s, scat t er di agr am, hi st ogr am, r un
char t , and cont r ol char t .
Step 3 Anal yz e the pr obl em.
Possi bl e t ool s i ncl ude Par et o char t , cause- and- ef f ect di agr am.
Step 4 Gener ate poten ti al sol uti ons.
M et hods i ncl ude br ai nst or mi ng, benchmar k i ng, i nt er vi ew i ng, and sur veyi ng.
Step 5 Choose a sol ut i on.
Be sur e w hat t he cr i t er i a ar e f or choosi ng a sol ut i on. ( Ref er t o t he goal est abl i shed i n St ep 1) . Appl y
cr i t er i a t o pot ent i al sol ut i ons and sel ect t he best one.
Step 6 I mpl ement the sol ut i on.
Keep ever yone i nf or med.
Step 7 M oni tor the sol uti on to see i f i t accompl i shes the goal .
I f not , modi f y t he sol ut i on or r et ur n t o St ep 1. Possi bl e t ool s i ncl ude cont r ol char t and r un char t .
Pr ocess I mpr ovement
A syst emat i c appr oach t hat i nvol ves document at i on, measur ement , and anal ysi s f or t he pur pose of i mpr ovi ng
t he f unct i on i ng of a pr ocess.
Typi cal goal s of pr ocess i mpr ovement i ncl ude: i ncr easi ng customer sati sf acti on , achi evi ng h i gher qual i ty,
r educi ng w aste, r educi ng cost, i ncr easi ng pr oducti vi ty, and speedi ng u p the pr ocess.
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 54
The Pl an- D o- Study- Act ( PD SA) Cycl e or t he D emi ng W heel i s t he concept ual basi s f or cont i nuous i mpr ovement
act i vi t i es.
Basi c St eps of t he PDSA Cycl e
Pl an . Begi n by st udyi ng t h e cur r en t pr ocess. D ocument
t hat pr ocess. Then col l ect dat a t o i dent i f y pr obl ems. Nex t ,
sur vey dat a and devel op a pl an f or i mpr ovement . Speci f y
measur es f or eval uat i ng t he pl an.
D o. I mpl ement t he pl an, on a smal l scal e i f possi bl e.
D ocument any changes made dur i ng t h i s phase. Col l ect dat a
syst emat i cal l y f or eval uat i on.
Study. Eval uat e t he dat a col l ect i on dur i ng t he do phase.
Check how cl osel y t he r esul t s mat ch t he or i gi nal goal s of t he
pl an phase.
Act . I f t he r esul t s ar e successf ul , st andar di ze t he new
met hod and communi cat e t he new met hod t o al l peopl e
associ at ed w i t h t he new pr ocess. I mpl ement t r ai ni ng f or t he
new met hod. I f t he r esu l t s ar e unsuccessf ul , r evi se t h e pl an and
r epeat t he cycl e, or cease t he pr oj ect .
Seven ( 7) Basi c Qual i t y Tool s
1 . Check Sheet
A t ool f or r ecor di ng and or gani zi ng dat a t o
i dent i f y a pr obl em.
For i dent i f yi ng t ypes of def ect s, w hen and
w her e t hey occur .
2 . Fl ow char t
A di agr am of t he st eps i n a pr ocess.
For i dent i f yi ng possi bl e poi n t s i n t he pr ocess
w her e pr obl ems occur .
3 . Scatter D i agr am
A gr aph t hat show s t he degr ee and di r ect i on
of r el at i onshi p bet w een t w o var i abl es.
M ay poi nt t o a cause of a pr obl em.
4 . H i stogr am
A char t of an empi r i cal f r equency
di st r i but i on.
Can be usef ul i n get t i ng a sense of t he
di st r i but i on of obser ved val ues; i f t he
di st r i but i on i s symmet r i cal , w hat t he r ange of
val ues i s, and i f t her e i s any unusual val ues.
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 55
5 . Par eto Anal ysi s
A t echni que f or cl assi f yi ng pr obl em ar eas accor di ng t o degr ee of i mpor t ance, and f ocusi ng on t he most
i mpor t ant .
6 . Cont r ol Char t
A st at i st i cal char t of t i me- or der ed val ues of a sampl e st at i st i c.
Can be used t o moni t or i f t he pr ocess out put i s i n cont r ol or out of cont r ol .
7 . Cause- and- ef f ect D i agr am
Al so cal l ed f i shbone or I shi k aw a di agr am.
Used t o sear ch f or t he cause( s) of a pr obl em.
SPC TO O LS CO M M O NLY USED FO R PRO BLEM SO LV I N G AND CO N TI N U O US I M PRO V EM EN T
1. Pr ocess Fl ow Char t A pi ct ur e w hi ch descr i bes t he mai n st eps, br anches and event ual out put s of a pr ocess
2. Par eto Anal ysi s A coor di nat ed appr oach f or i dent i f yi ng, r ank i ng, and w or k i ng t o per manent l y el i mi nat e def ect s.
Focuses on i mpor t ant er r or sour ces. 80/ 20 r ul e: 80 per cent of t he pr obl ems ar e due t o 20 per cent of t he causes.
A B C D F G
Fr equency Per cent age
100%
50%
0%
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 56
3. Run Char t A t i me sequence char t show i ng pl ot t ed val ues of a char act er i st i c.
4. H i stogr am A di st r i but i on show i ng t he f r equency of occur r ences bet w een t he hi gh and l ow r ange of data.
5. Scatter D i agr am Al so k now n as a cor r el at i on char t . A gr aph of t he val ue of one char act er i st i c ver sus anot her
char act er i st i c.
6. Check Sheet An or gani zed met hod f or r ecor di ng data.
ITEM S
_____
_____
A
B
C
D
E
F
G
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 57
7. Cause and Ef f ect D i agr am A t ool t hat uses a gr aphi cal descr i pt i on on t he pr ocess el ement s t o anal yze pot ent i al
sour ces of pr ocess var i at i on.
8. Cont r ol Char ts A t i me sequence char t show i ng pl ot t ed val ues of a st at i st i c, i ncl udi ng a cent r al l i ne and one or
mor e stat i st i cal l y der i ved cont r ol l i mi t s.
At t r i but e or var i abl e
Ti me
STATI STI CAL PROCESS CONTROL
A vi sual di spl ay cal l ed a cont r ol char t i s used to pl ot val ues of a measur e of pr ocess per f or mance ( e.g., t he t i me
di r ect or y assi st ance oper ator spends w i t h a cal l er ) t o det er mi ne i f t h e pr ocess i s i n cont r ol ( e.g., t he t i me i s l ess t han 30
seconds i n t he oper at or ex ampl e) .
The st eps i n const r uct i ng and usi ng qual i t y- cont r ol char t can be summar i zed as:
1. D eci de on some measur e of ser vi ce syst em per f or mance.
2. Col l ect r epr esent at i ve hi st or i cal dat a f r om w hi ch est i mat es of t he popul at i on mean and var i ance f or t he syst em
per f or mance measur e can be made.
3. D eci de on a sampl e si ze, and usi ng t he est i mat es of popul at i on mean and var i ance, cal cul at e ( by convent i on) 3
st andar d devi at i on cont r ol l i mi t s.
4. Gr aph t he cont r ol char t as a f unct i on of sampl e mean val ues ver sus t i me.
5. Pl ot sampl e means col l ect ed at r andom on t he char t , and i nt er pr et t he r esul t s as f ol l ow s:
a. Pr ocess i n cont r ol
b. Pr ocess out of cont r ol . I n t hi s case:
i . Eval uat e t he si t uat i on
i i . Tak e cor r ect i ve act i on
i i i . Check r esul t s of act i on
6. Updat e t he cont r ol char t on a per i odi c basi s, and i ncor por at e r ecent dat a.
Ef f ect
M achi ne
M at er i al M et hod
M an
Envi r onment
Pr ocess mean
LCL
UCL
LCL
Pr ocess-i n-cont r ol
Pr ocess-out -of -cont r ol
Pr ocess-out -of -cont r ol
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 58
Cont r ol char t s f or means f al l i nt o t w o cat egor i es based on t he t ype of per f or mance measur e. A var i abl e cont r ol char t
( X- char t ) r ecor ds measur ement s t hat per mi t f r act i onal val ues, such as l engt h, t i me, or w ei ght . An at t r i but e cont r ol chat
( p- char t ) r ecor ds di scr et e dat a, such as t he number of def ect s or er r or s as a per cent age.
St at i st i cal Pr ocess Cont r ol Pr obl ems
Sampl e si ze x - char t R- char t
n A2 D 3 D 4
2
3
4
5
5
7
8
9
10
12
14
16
18
20
22
24
1.880
1.023
0.577
0.729
0.483
0.419
0.373
0.337
0.308
0.266
0.235
0.212
0.194
0.180
0.167
0.157
0
0
0
0
0
0.076
0.136
0.184
0.223
0.283
0.328
0.363
0.391
0.415
0.434
0.451
3.267
2.574
2.282
2.114
2.004
1.924
1.864
1.816
1.777
1.717
1.672
1.637
1.608
1.585
1.566
1.548
Tabl e 1.1 Var i abl e Cont r ol Char t Const ant s
Tabl e 1.2 RI SK I N QUALI TY- CONTROL D ECI SI ONS
Q ual i ty- contr ol deci si on
Tr ue state of ser vi ce Take cor r ecti ve acti on D o nothi ng
Pr ocess i n cont r ol
Pr ocess out of cont r ol
Type I er r or ( pr oducer s r i sk )
Cor r ect deci si on
Cor r ect deci si on
Type I I er r or (consumer s r i sk )
Cont r ol char t i s a vi sual di spl ay t hat i s used t o pl ot val ues of a measur e of pr ocess per f or mance ( e.g., t he t i me a
di r ect or y assi st ance oper ator spends w i t h a cal l er ) t o det er mi ne i f t h e pr ocess i s i n cont r ol ( e.g., t he t i me i s l ess t han 30
seconds i n t he oper at or ex ampl e) . For ex ampl e, Fi gur e 1.1 show s a cont r ol char t t hat i s used t o moni t or emer gency
ambul ance r esponse t i me. Thi s char t i s a dai l y pl ot of mean r esponse t i me t hat per mi t s moni t or i ng per f or mance f or
unusual devi at i ons f r om t he nor m. W hen a measur ement f al l s out si de t he cont r ol l i mi t s t hat i s, above t he upper
cont r ol l i mi t ( UCL) or bel ow t he l ow er cont r ol l i mi t ( LCL) t he pr ocess i s consi der ed t o be out of cont r ol : consequent l y,
t he syst em i s i n need of at t ent i on.
Ex ampl e 1- 1: Cont r ol Char t f or Var i abl es ( x - char t )
The qual i t y- cont r ol char t f or mean ambul ance r esponse t i me show n i n Fi gur e 1.1 i s an ex ampl e of a var i abl e measur e.
I t i s based on t ak i ng a r andom sampl e of f our ambul ance cal l s each day t o cal cul at e a sampl e mean r esponse t i me t o
moni t or per f or mance. The r ange of t he sampl e val ues ( i .e., t he di f f er ence bet w een t he hi ghest and t he l ow est val ues)
w i l l be used i nst ead of t he st andar d devi at i on t o cal cul at e t he cont r ol l i mi t s. Assume t hat past r ecor ds of ambul ance
syst em per f or mance yi el d an est i mat ed popul at i on mean r esponse t i me of 5.0 mi nut es, w i t h an est i mat ed aver age r ange
of 3.1mi nut es. Appr opr i at e f or mul as f or cal cul at i ng t he cont r ol l i mi t s f or an x - char t usi ng t he R r ange as a subst i t ut e
f or t he st andar d er r or of t he mean ar e
UCL = x + A2 R
LCL = x + A2 R
w her e x = est i mat e of popul at i on mean
R = est i mat e of popul at i on r ange
A2 = val ue f r om Tabl e 1.2 f or sampl e si ze n
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 59
For ambul ance r esponse t i me cont r ol char t show n i n Fi gur e 1.1, t he cont r ol l i mi t s f or a sampl e si ze of f our ar e
cal cul at ed as f ol l ow s:
UCL = x + A2 R = 5.0 + ( 0.729) ( 3.1) = 7.26
LCL = x + A2 R = 5.0 - ( 0.729) ( 3.1) = 2.74
W e al so coul d moni t or t he var i ance of t he ambul ance r esponse t i mes usi ng t he r ange ( i .e., t he l ar gest val ue mi nus t he
smal l est val ue i n t he sampl e) as a measur e of var i ance. A r ange or R- char t can be const r uct ed i n a manner si mi l ar t o
t he x - char t by usi ng t he f ol l ow i ng f or mul as:
UCL = D 4 R
LCL = D 3 R
w her e R = est i mat e of popul at i on r ange
D 4 = UCL val ue f r om Tabl e 1.2 f or sampl e si ze n
D 3 = LCL val ue f r om Tabl e 1.2 f or sampl e si ze n
For our ambul ance case, t he r ange cont r ol l i mi t s ar e cal cul at ed usi ng t he constant s i n Tabl e 1.2 f or a sampl e si ze of
f our :
UCL = D 4 R = ( 2.282) ( 3.1 ) = 7.1
LCL = D 3 R = ( 0) ( 3.1) = 0
M
e
a
n
r
e
s
p
o
n
s
e
t
i
m
e
,
m
i
n
u
t
e
s
7.26
5.0
2.74
x
x
x
x
x
x x
x
Upper cont r ol l i mi t ( UCL)
x
Low er cont r ol l i mi t ( LCL)
1 2 3 4 5 6 7 8
Ti me, days
Fi gur e 1.1 x - char t f or ambul ance r esponse.
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 60
Name: D at e:
Cour se & Year : Schedul e:
Pr obl em Set f or St at i st i cal Pr ocess Cont r ol
1 . Con t r ol Ch ar t f or Var i abl e
To be come pr odu ct i ve, Resor t I n t er nat i on al i s i n t er est ed i n set t i n g st an dar ds f or t h e t i me t hat t el eph on e r eser vat i on cl er k s
spen d w i t h vacat i on er s mak i n g tou r ar r an gemen t s. Col l ect i n g dat a on t h e amou n t of t i me t h e r eser vat i on cl er k spen d w i t h
cu st omer s h as been pr oposed to det er mi n e t h e mean t i me an d aver age r an ge as w el l as to est abl i sh a pr ocess con tr ol char t f or
t h i s oper at i on . Th e t abl e bel ow r ecor ds t h e t i me i n mi n u t es t h at r eser vat i on cl er k s spen t an sw er i n g cal l s as f ou n d by obser vi n g
on e cal l each day dur i n g a t ypi cal w eek . Th e f i f t h r ow con tai n s t h e X val u es f or each day. Th e l ast r ow con tai n s t h e r ange ( i .e.,
h i gh - l ow ) val u es f or each day ( e.g., t h e h i gh f or M on day w as 1 4 an d t h e l ow w as 5 , yi el di n g a r an ge of 9 ) . Compu t e f or a)
popu l at i on mean an d r an ge, b) est abl i sh t h e con t r ol l i mi t s f or an x - char t w h en sampl i n g f our r andom cal l s each day f or each
cl er k , c) con st r uct an R- char t , an d d) pl ot t h e aver age cal l t i me.
Cl er k M on day Tu esday W edn esday Th u r sday Fr i day
Al i ce
Bi l l
Jan i ce
M i k e
5
6
1 4
8
1 1
5
1 3
6
1 2
1 2
1 0
9
1 3
1 0
9
1 2
1 0
1 3
9
1 4
x
Ran ge
8 .2 5
9
8 .7 5
8
1 0 .7 5
3
1 1 .0
4
1 1 .5
5
2 . Contr ol Char t f or Attr i bute
A r egi onal ai r l i n e i s concer n ed abou t i t s r ecor d of on - t i me per f or mance. Th e M emph i s h u b ex per i en ce 2 0 f l i gh t oper at i on s
each day of t h e w eek , w i t h t h e f ol l ow i n g r ecor d of on- t i me depar t u r es f or t h e pr evi ou s 1 0 days: 1 7 , 1 6 , 1 8 , 1 9 , 1 6 , 1 5 , 2 0 , 1 7 ,
1 8 , an d 1 6 . Pr epar e a p- char t w i t h a sampl e con si st i n g of 1 w eek s aver age on- t i me depar t ur e per cen t age.
3 . Th e t i me to mak e beds at a mot el sh ou l d f al l i n to an agr eed- on r an ge of t i mes. A sampl e of f our mai ds w as sel ect ed, an d t h e
t i me n eeded t o mak e a bed w as obser ved on t h r ee di f f er en t occasi on s:
Ser vi ce ti m e, seconds
M ai d Sam pl e 1 Sam pl e 2 Sam pl e 3
An n
Li n da
M ar i e
M i ch ael
1 2 0
1 3 0
2 0 0
1 6 5
9 0
1 1 0
1 8 0
1 5 5
1 5 0
1 4 0
1 7 5
1 4 0
a. D et er mi n e t h e u pper l i mi t and l ow er l i mi t f or an X- char t an d an R- ch ar t w i t h a sampl e si ze of f ou r .
b. Af t er t h e con t r ol char t w as est abl i sh ed, a sampl e of f our obser vat i on s had t h e f ol l ow i n g t i mes i n secon ds: 1 8 5 , 1 5 0 , 1 92 ,
an d 1 7 8 . I s cor r ect i ve act i on n eeded?
4 . Th e man agemen t of t h e D i n er s D el i gh t f r anch i sed r est aur an t ch ai n i s i n t h e pr ocess of est abl i sh i n g qu al i t y- con tr ol ch ar t f or
t h e t i me t h at each cu st omer i s gi ven to each cu st omer . M anagemen t t h i n k s t h e l en gt h of t i me t hat each cu st omer i s gi ven
sh ou l d r emai n w i t h i n cer t ai n l i mi t s to en han ce ser vi ce qual i t y. A sampl e of si x ser vi ce peopl e w as sel ect ed, an d t h e cust omer
ser vi ce t h ey pr ovi de w as obser ved f ou r t i mes. Th e act i vi t i es t h at t h e ser vi ce peopl e w er e per f or mi n g w er e i den t i f i ed, and t h e
t i me t o ser vi ce on e cu stomer w as r ecor ded as not ed bel ow :
Ser vi ce ti m e, seconds
Ser vi ce per son
Sam pl e 1 Sam pl e 2 Sam pl e 3 Sam pl e 4
1
2
3
4
5
6
2 0 0
1 2 0
8 3
6 8
1 1 0
1 1 5
1 5 0
8 5
9 3
1 5 0
9 0
6 5
1 7 5
1 0 5
1 3 0
1 4 5
7 5
1 1 5
9 0
7 5
1 5 0
1 7 5
1 0 5
1 2 5
a. D et er mi n e t h e u pper and l ow er con t r ol l i mi t s f or an X- char t an d an R- ch ar t w i t h a sampl e si ze of 6 .
b. Af t er t h e con t r ol char t w as est abl i sh ed, a sampl e of si x ser vi ce peopl e w as obser ved an d t h e f ol l ow i n g cu stomer ser vi ce
t i mes i n secon ds w er e r ecor ded: 1 8 0 , 1 2 5 , 1 1 0 , 9 8 , 1 5 6 , and 1 9 0 . I s cor r ect i ve act i on cal l ed f or ?
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 61
TIP TOP MARKET
Tip Top Market is a regional chain of supermarkets located in the Southeastern United States. Karen Martin, manager of one of the
stores, was disturbed by the large number of complaints from customers at her store, particularly on Tuesdays, so she obtained
complaint records form the stores customer service desk for the last eight Tuesdays.
Assume you have been asked to help analyzed the data and to make recommendations for improvement. Analyze the data using a
check sheet, a Pareto diagram, and run charts. Then construct a cause-and-effect diagram for the leading category on your Pareto
diagram.
On July 15, changes were implemented to reduce out-of-stock complaints, improve store maintenance, and reduce checkout line or
pricing problems. Do the results of the last two weeks reflect improvement?
Base on your analysis; prepare a list of recommendations that will address customer complaints.
June 1
out of orange yogurt
bread stale
checkout lines too long
overcharged
double charged
meat smelled strange
charged for item not purchased
couldnt find the sponges
meat tasted strange
store too cold
light out in parking lot
produce not fresh
lemon yogurt pass sell date
couldnt find rice
milk past sell date
stock clerk rude
cashier not friendly
out of maple walnut ice cream
something green in meat
didnt like music
checkout line too slow
June 8
fish smelled funny
out of diet bread
dented can
out of hamburger rolls
fish not fresh
cashier not helpful
meat tasted bad
ATM ate card
slippery floor
music too loud
undercharged
out of roses
meat spoiled
overcharged on two items
store too warm
out of rice
telephone out of order
overcharged
rolls stale
bread past sale date
June 15
wanted smaller size
too cold in store
out of Wheaties
out of Minute Rice
cashier rude
fish tasted fishy
ice cream thawed
double charged on hard rolls
long wait at check out
wrong price on item
overcharged
fish didnt smell right
overcharged on special
couldnt find aspirin
undercharged
checkout lines too long
out of diet cola
meat smelled bad
overcharged on eggs
bread not fresh
didnt like music
lost wallet
overcharged on bread
June 22
milk past sales date
store too warm
foreign object in meat
store too cold
eggs cracked
couldnt find lard
out of 42 oz. Tide
fish really bad
windows dirty
couldnt find oatmeal
out of Bounty paper towels
overcharged on orange juice
lines too long at checkout
couldnt find shoelaces
out of Smuckers strawberry jam
out of Frosty Flakes cereals
out of Thomas English Muffins
June 29
checkout line too long
out of Dove soap
out of Bisquick
eggs cracked
store not clean
store too cold
cashier to slow
out of skim milk
charged wrong price
restroom not clean
couldnt find sponges
checkout lines slow
out of 18 oz. Tide
out of Campbells turkey soup
out of pepperoni sticks
checkout lines too long
meat not fresh
overcharged on melon
July 6
out of straws
out of bird food
overcharged on butter
out of masking tape
stockboy was helpful
lost child
meat looked bad
overcharged on butter
out of Swiss chard
too many people in store
out of bubble bath
out of Dial soap
store too warm
price not as advertised
need to open more checkout
shopping carts hard to steer
debris in aisles
out of Drano
out of Chinese cabbage
store too warm
floors dirty and sticky
out of Diamond chopped walnuts
July 13
wrong price on spaghetti
water on floor
store looked messy
store too warm
checkout lines too long
cashier not friendly
out of Cheese Doodles
triple charged
out of Saran Wrap
out of Dove Bars
undercharged
out of brown rice
out of mushrooms
overcharged
checkout wait too long
shopping cart broken
couldnt find aspirin
out of Tip Top lunch bags
out of Tip Top straws
July 20
out of cucumbers
checkout lines too slow
found keys in parking lot
lost keys
wrong price on sale item
overcharged on corn
wrong price on baby food
out of 18 oz. Tide
out of Tip Top tissues
checkout lines too long
out of romaine lettuce
out of Tip Top toilet paper
out of red peppers
out of Tip Top napkins
out of apricots
telephone out of order
out of cocktail sauce
water on floor
out of onions
out of squash
out of iceberg lettuce
out of Tip Top paper towels
July 27
out of bananas
reported accident in parking lot
wrong price of cranapple juice
out of carrots
out of fresh figs
out of Tip Top napkins
out of Tip Top straws
windows dirty
out of iceberg lettuce
dislike store decorations
out of Tip Top lunch bags
out of vanilla soy milk
wanted to know who won the
lottery
store too warm
oatmeal spilled in bulk section
telephone out of order
out of Tip Top tissues
water on floor
out of Tip Top paper towels
out of Tip Top toilet paper
spaghetti sauce on floor
out of Peter Pan crunchy peanut
butter
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 62
Template of Quality Tools Required in the Case:
A. Check sheet
Complaint
category
Category description Frequency of occurrence
A Out of stock merchandise
B Expired or spoiled stock
C Cashier related complaints lines, charging
D Missing stocks
E Undesirable store ambiance temperature, lighting
F Poor customer service by other personnel
G Equipment related complaints
B. Pareto diagram (based on check sheet data)
C. Run chart (only for the category with the highest frequency of occurrences)
D. Control chart (only for the category with the highest frequency of occurrences)
E. Cause-and-effect diagram (only for the category with the highest frequency of occurrences)
F. List of recommendations (five alternative course of actions to address the most probable cause of the complain
category with the highest frequency of occurrence).
(Nota Bene: Group output should be written in intermediate paper. List the group members names that are only present)
C
u
m
u
l
a
t
i
v
e
f
r
e
q
u
e
n
c
y
o
f
o
c
c
u
r
r
e
n
c
e
F
r
e
q
u
e
n
c
y
o
f
o
c
c
u
r
r
e
n
c
e
A B C D E F (For illustr ation only )
100 %
F
r
e
q
u
e
n
c
y
o
f
o
c
c
u
r
r
e
n
c
e
6/ 1
(For illustr ation only )
0
6/ 8 6/ 15 6/ 222 6/ 29 7/ 6
7/ 13 7/ 20 (time)
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 63
F. PLANNING & DESIGNING OF P/O SYSTEM - Part II: FACILITY LAYOUT DESIGN
Lear ni ng Obj ect i ves
At t he end of t hi s modul e mat er i al , st udent shoul d be abl e t o;
1. D et er mi ne f aci l i t y l ayout desi gns
2. Li st t he advant ages of good l ayout
3. Recogni ze t he consi der at i ons of a good l ayout
4. D escr i be t he basi c t ypes of l ayout s
5. D evel op si mpl e pr ocess l ayout
6. Sol ve si mpl e l i ne bal anci ng pr obl em
7. Sol ve l oad di st ance pr obl em
Faci l i ty Layout
Faci l i t y l ayout i s t he desi gni ng of .. or conf i gur at i on of depar t ment s, .., and equ i pmen t t hat const i t u t e t he
.. pr ocess t o enhance t he of t he syst em and smoot hen t he . of t he pr oduct and/ or
ser vi ce.
Advantage of a Good Layout
1. Few er ..
2. .. mat er i al f l ow
3. Bet t er .
4. Reduce mat er i al handl i ng ..
5. Reduce .
6. Ef f ect i ve .. ut i l i zat i on
Consi der at i on of a Good Layout
1. .. and .. r equi r ement s. Capaci t y deci si ons ar e t he pr e- r equ i si t e t o a good l ayout . Onl y w hen w e k now
t he per sonnel , machi nes, and equi pment r equi r ed t hat w e can pr oceed w i t h t he l ayout and pr ovi de space f or each
component s of t he conver si on syst em.
2. Envi r onment al and .. D eci si ons may be r equi r ed about w i ndow s, pl ant er s, and hei ght of par t i t i ons t o
f aci l i t at e ai r f l ow or .. t o r educe noi se, ., t o pr ovi de pr i vacy, and so on.
3. Fl ow s of .. D eci si ons about t he best w ay t o f aci l i t at e communi cat i ons must be al so made. Thi s may r equi r e
deci si on about pr ox i mi t y as w el l as open .. ver sus hal f - hei ght s di vi der s ver sus pr i vat e of f i ces.
4. Cost of .. bet w een var i ous ar eas. Ther e may be uni que consi der at i ons r el at ed t o t he di f f i cu l t y of movement s
or t he i mpor t ance of cer t ai n ar eas bei ng adj acent t o each ot her .
5. .. equi pment . D eci si ons ar e al so r equi r ed about equi pmen t t o be used; such as conveyor s,
cr anes, f or k l i f t s, and car t s t o del i ver mat er i al or mai l cor r espondences.
Type of M anuf actur i ng and Ser vi ce O per ati ons
1. I nt er mi t t ent oper at i ons i ncl ude conver si on syst em t hat pr oduced .. vol umes, but .......... var i et y of i t ems or
ser vi ces ( such as: .., , and so on) .
2. Cont i nuous oper at i ons composed of conver si on syst em t hat pr oduced .. of one or f ew st an dar di ze i t ems;
ex ampl es ar e , .., and
3. Pr oj ect i s used t o pl an and coor di nat e .. j obs w i t h r el at i vel y l i f e span f or i nst ance .,
.., and ..
Si mpl e Pr ocess Layout
A C E
1 2 3
B D F
4 5 6
Locat i on W or k stat i on t o be assi gned
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 64
D esi gni ng Pr oduct Layout
Li ne Bal anci ng
Li ne bal anci ng ( Yamazumi ) i s t he pr ocess of assi gni ng .. t o .. t hat have appr ox i mat el y equal t i me
r equi r ement s.
Cycl e t i me i s t he .. t i me al l ow ed at each w or k stat i on t o compl et e i t s set s of task or a uni t .
H er e i s a si mpl e def i n i t i on and ex ampl e of l i ne bal anci ng:
Ever yone i s doi ng t he same amount of w or k
D oi ng t he same amount of w or k t o cust omer r equi r ement
Var i at i on i s smoot hed
No one over bur dened
No one w ai t i ng
Ever yone w or k i ng t oget her i n a BALAN CED f ashi on
Her e w e see oper at or number 1 over - pr oduci ng, t hus
cr eat i ng t he ot her 6 w ast es
W e si mpl y r e- bal ance t he w or k cont ent ( Re- di st r i but e some
of t he w or k ) , Usi ng a Yamazumi boar d as i s of t en k now n.
I n t he ex ampl e show n above, t he Yamazumi ex er ci se
l ook s i ncr edi bl y si mpl e so w hy doesn t ever y do i t ?
Ther e ar e t w o vi t al pr e- r equi si t e t o bal anci ng a l i ne w hi ch must be i n pl ace bef or e a Yamazumi ex er ci se can t ak e pl ace.
These ar e Tak t Ti me and St andar d W or k
Tak t Ti me
Tak t Ti me comes f r om a Ger man w or d t ak t meani ng r hyt hm or beat . I t i s a t er m of t en associ at ed w i t h t he t ak t t he
conduct or set s so t hat t he or chest r a pl ays i n uni son. Tak t Ti me i s used t o mat ch t he pace of w or k t o t he aver age pace of
cust omer demand. Tak t i s not a number t hat can be measur ed ( M i sconcept i on # 1) , and i s not t o be mi st ak en w i t h Cycl e
Ti me, w hi ch i s t he t i me i t t ak es t o compl et e one t ask . Cycl e Ti me may be l ess t han, mor e t han, or equal t o Tak t Ti me.
You can never measur e Tak t Ti me w i t h a st op w at ch. You must cal cul at e i t . The f or mul a f or Tak t Ti me i s:
Tak t Ti me =
Net Avai l abl e Ti me per Day
Cust omer D emand per Day
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 65
Tak t Ti me i s ex pr essed as seconds per pi ece , i ndi cat i ng t hat cust omer s ar e buyi ng a pr oduct once ever y so many
seconds. Tak t Ti me i s not ex pr essed as pi eces per second ( M i sconcept i on # 2) . By paci ng pr oduct i on to t hi s r at e of
cust omer demand, Lean M anuf act ur i ng seek s t o mi ni mi ze w as and ensur e on- t i me at a l ow cost .
Tak t t i me i s t he pr oduct i on D r umbeat bases on cust omer demand.
St andar d W or k
W het her you use st andar d w or k combi nat i on t abl es, st andar d w or k i nst r uct i on sheet s or any ot her st andar d w or k
document at i on w i l l depend upon t he t ype of w or k i nvol ved.
St andar d w or k i nst r uct i on sheet s pr ovi de a det ai l ed descr i pt i on of HOW t o do a par t i cul ar st ep of a w or k oper at i on.
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 66
Benef i t s of t he pr e- r equi si t es:
Takt ti me max i mi z es the pr oducti vi ty due to: Standar d oper ati ons pr ovi des:
1. Easi l y managed pr ocesses
2 . Out put of each pr ocess mat ches cust omer
demand
1. Capabl e and r epeat abl e pr ocesses
2. Pr ocess cont r ol at sour ce
3. I mpr oves accur acy of pl anni ng
4. Bet t er adher ence t o pl ans
5. A pl at f or m f r om w hi ch cont i nuous
i mpr ovement can be made
6. Reduce cost s
h t t p:/ / w w w .beyon dl ean .com/ l i n e- bal an ci n g.h t ml dat e r et r i eved 0 4 M ay 2 0 1 1 2 1 :5 4 :2 7
A Yamazumi char t i s a stack ed bar char t t hat show s t he bal ance of cycl e t i me w or k l oads bet w een a number of
oper at or s typi cal l y i n an assembl y l i ne or w or k cel l . The Yamazumi char t can be ei t her f or a si ngl e pr oduct or mul t i
pr oduct assembl y l i ne.
Yamazumi i s a Japanese w or d t hat l i t er al l y means t o st ack up.
Toyot a uses Yamazumi w or k bal ance char t s t o vi sual l y pr esent t he w or k cont ent of a ser i es of task s and f aci l i t at e w or k
bal anci ng and t he i sol at i on and el i mi nat i on of non val ue added w or k cont ent .
Sour ce: ht t p:/ / w w w .acsco.com/ Yamazumi .ht m dat e r et r i eved 04 M ay 2011 21:59:24
SOM E USEFUL FORM ULA:
O ut put r ate =
Oper at i ng t i me per day
Cycl e t i me
The t ask t i mes gover n t he r ange of possi bl e cycl e t i mes. The .. cycl e t i me i s equal t o t he . t ask t i me, and
t he max i mum cycl e t i me i s equal to t he sum of t he t ask t i mes.
Cycl e ti me =
Oper at i ng t i me per day
D esi r e out put r at e
As a gener al r ul e, t he cycl e t i me i s det er mi ned by t he desi r ed out put ; t hat i s, a desi r ed out put l evel i s sel ect ed, and t he
cycl e t i me
Theor et i cal mi ni mum number of stati ons =
Sum of t ask t i mes
Cycl e t i me
Pr ecedence di agr am i s a di agr am t hat show s t he el ement al t ask s and t hei r pr ecedence r equi r ement s.
Tw o w i del y used measur e of ef f ect i veness ar e
1. The per cent age of i dl e t i me of t he l i ne. Thi s i s somet i mes r ef er r ed t o as t he bal ance del ay. I t can be comput ed as
f ol l ow s;
Per cen tage of i dl e ti me =
I dl e t i me per cycl e
100
Act ual number of w or kst at i ons Cycl e t i me
I n ef f ect , t hi s i s t he aver age i dl e t i me di vi ded by t he cycl e t i me, mul t i pl i ed by 100. Not e t hat cycl e t i me r ef er s t o t he
act ual cycl e t i me t hat i s achi eved.
2. The ef f i ci ency of t he l i ne. Thi s i s comput ed as f ol l ow s
Ef f i ci ency = 100 Per cent age of i dl e t i me
Sampl e Pr obl em:
Task I mmedi ate f ol l ow er Task ti me ( i n mi nut es)
a
b
c
d
e
f
g
h
b
e
d
f
f
g
h
end
0.2
0.2
0.8
0.6
0.3
1.0
0.4
0.3
t = 3.8
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 67
Usi ng t he i nf or mat i on cont ai ned i n t he t abl e show n, do each of t he f ol l ow i ng:
1. D r aw a pr ecedence di agr am.
2. Assumi ng an ei ght - hour w or k day, comput e t he cycl e ti me needed t o obt ai n an out put of 400 uni t s per day.
3. D et er mi ne t he mi ni mum number of w or k stat i ons r equi r ed.
4. Assi gn t ask t o w or k st at i on usi ng t hi s r ul e: Assi gn t ask s accor di ng t o gr eat est number of f ol l ow i ng t ask s. I n case of a
t i e, use t he t i ebr eak er of assi gni ng t he t ask w i t h t he l ongest pr ocessi on t i me f i r st .
5. Comput e t he r esul t i ng per cent i dl e t i me and ef f i ci ency of t he syst em.
Pr ecedence di agr am:
Cycl e t i me
CT =
OT
=
480 mi nu t es per day
= 1.2 mi nut es per day
D 400 uni t s per day
Number of w or k st at i on
Nmi n =
Dt
=
400 uni t s per day 3.8 mi nut es
= 3.17 w or k st at i ons
OT 480 mi nu t es per day
Schedul e of i dl e t i me
Stati on
Ti me
r emai n i ng El i gi bl e W i l l f i t
Assi gn ( task
t i me) I dl e t i me
1
2
3
4
1.2
1.0
0.2
0.0
1.2
0.6
0.3
1.2
0.2
1.2
0.8
0.5
a,c
c,b
b,d
e,d
e,d
e
f
f
g
g
h
-
a,c
c,b
b
none
e.d
e
none
f
none
g
h
-
a( 0.2)
c( 0.8)
b( 0.2)
-
d( 0.6)
e( 0.3)
-
f ( 1.0)
-
g( 0.4)
h( 0.3)
-
0.0
0.3
0.2
0.5
1 . 0
Per cent age of i dl e t i me
% I T =
I T
=
1.0 mi nut e per cycl e
= 0.2083
Nact u al CT 4 st at i ons 1.2 mi nut es per cycl e
Ef f i ci ency of t he l i ne bal ance
Ef f i ci ency = 100 % - per cent of i dl e t i me
Ef f i ci ency = 100 - 20.83 = 79 %
a b
c d
e
f g h
0.2 0.2 0.3
0.8
0.6
1.0 0.4
0.3
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 68
Cl oseness Rat i ng
A
E
I
O
U
X
: Absol ut el y necessar y
: Especi al l y i mpor t ant
: I mpor t ant
: Or di nar y cl oseness OK
: Uni mpor t ant
: Undesi r abl e
Assi gnment of t he cl oseness r at i ng i s
subj ect i ve.
Rul e of t humb:
Ver y f ew A and X r el at i onshi ps shoul d
be assi gned. ( no mor e t han 5% of t he
cl oseness r at i ngs t o be an A and X) .
No mor e t han 10% shoul d be an E.
No mor e t han 15% t o be an I .
No mor e t han 20% t o be an O.
Thi s means t hat about 50% of t he
r el at i onshi ps shoul d be U.
Sampl e Pr obl em f or Cl oseness Rat i ng
Cl oseness r at i ng
1. Ar r ange si x depar t ment s i nt o a 2 x 3 gr i d so t hat
t hese condi t i ons ar e sat i sf i ed: 1 cl ose t o 2, 5 cl ose
t o 2 and 6, 2 cl ose t o 5, and 3 not cl ose t o 1 or 2.
2. Usi ng t he i nf or mat i on gi ven i n pr ecedi ng
pr obl ems, devel op a M ut her - t ype gr i d usi ng t he
l et t er s A, O, and X. Assume t hat any pai r of
combi nat i on not ment i oned have an O r at i ng.
Load D i st ance Techni que
M i ni mi zi ng Tr anspor t at i on Cost s or D i st ance
Assumpt i on: Cost s ar e di r ect , l i near f unct i on of di st ance.
Sampl e Pr obl em
Assi gn t he t hr ee depar t ment s t o l ocat i on A, B, and C, w hi ch ar e separ at ed by t he di st ance sh ow n i n Tabl e 2, i n such a
w ay t hat t r anspor t at i on cost i s mi ni mi zed. I f i t cost per met er t o move any l oad i s P5, comput e f or t he dai l y
t r anspor t at i on cost
Use t hi s heur i st i c r ul e: Assi gn depar t ment s w i t h t he gr eat est i nt er depar t ment al w or k f l ow f i r st .
Ri char d Mut her devel oped a mor e gener al appr oach t o t he pr obl em, whi ch al l ows f or
subj ect i ve i nput fr om anal yst s or manager s t o i ndi cat e t he r el at i ve i mpor t ance of
each combi nat i on of depar t ment pai r s. I n pr act i ce, t he l et t er s on t he gr i d ar e of t en
accompani ed by number s t hat i ndi cat e t he r eason f or each assi gnment ; t hey ar e
omi t t ed her e t o si mpl if y t he i l l ust r at i ons. Mut her suggest s t he f ol l owi ng l i st .
1. Use same equi pment or f aci l i t i es
2. Shar e t he same per sonnel or r ecor ds
3. Sequence of wor k f l ow
4. Ease of communi cat i on
5. Unsaf e or unpl easant condi t i ons
6. Si mi l ar wor k per f ormed
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 69
Tabl e 1. Locat i on ( D i st ance)
To
Fr om
A B C
A - 20 40
B 20 - 30
C 40 30 -
Tabl e 2. D epar t ment ( Load)
To
Fr om
1 2 3
1 - 10 80
2 20 - 30
3 90 70 -
Repr esent at i on :
Sol ut i on:
D epar t ment Number of Loads t o Locat i on D i st ance ( met er s) t o Load D i st ance Cost
1
2
3
2:10
3:80
1:20
3:30
1:90
2:70
A
C
B
C:40
B:20
A:40
B:30
A:20
C:30
400
1600
800
900
1800
2100
7600
P38000
Tr anspor t at i on M odel
The t r anspor t at i on pr obl em i nvol ves f i ndi ng t he l ow est - cost pl an
f or di st r i bu t i ng st ock s of goods or suppl i es f r om mul t i pl e
dest i nat i ons t hat demand t he goods or suppl i es. The t r anspor t at i on
model can be used t o det er mi ne how t o al l ocat e t he suppl i es
avai l abl e f r om t he var i ous f act or i es t o t he w ar ehouse t hat st ock or
demand t hose goods, i n such a w ay t hat t ot al shi ppi n g cost ( t i me,
di st ance) i s mi ni mi zed.
The i nf or mat i on needed t o use t he model consi st s of t he f ol l ow i ng:
1. A l i st of t he or i gi ns and each one s capaci t y or suppl y
quant i t y per per i od.
2. A l i st of t he dest i nat i ons and each one s demand per i od.
3. The uni t cost of shi ppi ng i t ems f r om each or i gi n t o each
dest i nat i on.
Tr anspor t at i on model must sat i sf y t hese assumpt i ons:
1. The i t ems t o be shi pped ar e homogeneous ( i .e., t hey ar e t he same r egar dl ess of t hei r sour ce of dest i nat i on) ,
2. Shi ppi ng cost per uni t i s t he same r egar dl ess of t he number of uni t s shi pped, and
3. Ther e i s onl y one r out e or mode of t r anspor t at i on bei ng used bet w een each sour ce and each dest i nat i on.
The descr i pt i on of t he t echni que of t he maj or st eps i n t he pr ocess i n t hi s or der :
1. Obt ai n an i ni t i al sol ut i on ( usi ng i nt ui t i ve l ow est cost appr oach or gr eedy met hod, nor t hw est t echni que, an d so
on)
2. Test i ng t he sol ut i on of opt i mal i t y
a. Eval uat i ng empt y cel l : St eppi ng- st one met hod
b. Eval uat i ng empt y cel l : The M odi f i ed D i st r i but i on met hod
3. I mpr ovi ng subopt i mal sol ut i ons
2
1 3 B
A
C
40m
20m
30m
Analysis:
Department
Workflow
(Load) Trip
Distance
(meters)
3-1
1-3
3-2
2-3
2-1
1-2
90
80
70
30
20
10
A-B
B-A
B-C
C-B
A-C
C-A
20
20
30
30
40
40
Sample Transportation Problem
Warehouse
Factory
A B C D
Supply
1
4 7 7 1
100
2
12 3 8 8
200
3
8 10 16 5
150
Demand 80 90 120 160
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 70
WAI T I N G LI N E M AN AGE M E N T
Thi s Book of t he Law shal l not depar t f r om your mout h, but you shal l medi t at e i n i t day and ni ght , t hat you may
obser ve t o do accor di ng t o al l t hat i s w r i t t en i n i t . Joshua 1:8
Le a rning O bje c tive s
1. D i scuss the basi c el ements of wai ti ng li ne probl ems
2. Provi de standard steady-state f ormula f or sol vi ng wai ti ng l i ne probl ems
Economics of the Waiting Line Problem
1. Cost Effect i veness Balance. I ni ti al l y, wi th mi ni mal servi ce capaci ty, the wai ti ng l i ne cost i s at a maxi mum. As servi ce
capaci ty i ncreased, there i s a reducti on i n the number of customer i n the l i ne and i n thei r wai ti ng ti mes, whi ch decrease
wai ti ng l i ne costs.
2. Pract i cal Vi ew of Wai t i ng Li nes. One i mportant variabl e is the number of arrival s over the course of the hours that
the servi ce system i s open. From the servi ce deli very vi ewpoi nt, customers demand varyi ng amounts of servi ce, of ten
exceedi ng normal capaci ty. We can have some control over arri val s i n vari ety of ways. For exampl e, we can have a short
l i ne (such as a drive-i n at a f ast-f ood restaurant wi th onl y several spaces), we can establi sh speci f i c hours f or speci f ic
customers, or we can run speci al s. For the server, we can af f ect servi ce ti me by usi ng f aster or sl ower server, f aster or
sl ower machi nes, di f f erent tool i ng, di f f erent materi al , di f f erent l ayout, f aster setup ti me, and so on. The essenti al poi nt i s
wai ti ng l i ne are not a f i xed condi ti on of a productive system but are to a very l arge extent wi thi n the control of the system
management desi gn.
The Queuing System
The queui ng system consi sts of essenti all y of three maj or components:
The source of populat i on and t he way cust omers arri ve at t he syst em
Cust omer Arri val : Arrival at a servi ce system may be drawn f rom a f i ni te or an i nf i ni te
1. Popul at ion Source
a. Fi ni t e Source ref ers to the l i mi ted-si ze customer pool that wil l use the servi ce and, at ti mes, f orm a l i ne. The reason
f or thi s f i ni te cl assif i cati on i s i mportant because when a customer l eaves i ts posi ti on as a member f or the popul ati on
(a machi ne breaki ng down and requi ri ng servi ce, f or exampl e), the si ze of the user group is reduced by one, whi ch
reduces the probabi l i ty of the next occurrence. Conversel y, when a customer i s servi ced and returns to the user group,
the popul ati on i ncreases and the probabi l i ty of a user requi ri ng servi ce al so i ncreases. Thi s f i ni te cl ass of probl ems
requi res a separate set of f ormul as f rom that of the i nf i ni te popul ati on case.
b. I nfi ni t e Popul at i on i s one l arge enough i n relati on to the servi ce system so that the popul ati on si ze caused by
subtracti on or addi ti on to the popul ati on (a customer needi ng servi ce customer returni ng to the popul ati on) does not
si gnif i cantl y af f ect the system probabi l i ti es.
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 71
2. Di st ri but i on of Arri val s. When descri bi ng a wai ti ng system, we need to def i ne the manner i n whi ch customer or the
wai ti ng uni ts are arranged f or servi ce.
a. Arri val rat e. Wai ti ng li ne f ormulas generall y requi re and arri val rate, or the number of uni ts per peri od (such as an
average of one every si x mi nutes). A constant arri val di stri buti on i s peri odi c, wi th exactl y the same ti me peri od
between successive arri val s. I n producti ve system, about the onl y arri val s that trul y approach a constant i nterval
peri od are those that are subj ect to machi ne control . Much more common vari abl e (random) arri val di stri buti on.
b. Vi ewpoi nt s in observi ng arri val s at a servi ce faci l i t y
i. Fi rst, we can anal yze the ti me between successive arri val s to see i f the ti mes f ol l ow some statisti cal di stri buti on.
Usuall y, we assume that the ti me between arri val s is exponenti al l y di stri buted.
Exponent i al Di st ri but ion. I n the f i rst case, when arrival s at a servi ce f aci li ty occur i n a purel y random f ashi on,
a pl ot of the i nter-arri val ti mes yi el ds an exponenti al di stributi on. The probabi l i ty f uncti on i s:
(t) = e-
t
where i s the mean number of arri vals per ti me peri od; e i s the natural al gori thm (2.7183), and t i s the arri val
ti me.
ii. Second, we can set some ti me l ength (T) and try to determi ne how many arri vals might enter the system wi thi n T.
We typi call y assume that the number of arri vals per ti me is Poi sson di stri buted.
P
T
(n) =
(T)
n
e-
t
n!
Poi sson i s a discrete di stri buti on (smoothed curve, the curve become smoother as n becomes l arge) because n
ref ers to the number of arri val i n a system, and thi s must be an i nteger. (For exampl e, there cannot be 1.5 arri val s).
Al so not that the exponenti al and Poi sson di stri buti on can be deri ved f rom one another, the mean and vari ance of
the Poi sson di stri buti on are equal and denote by . The mean of the exponenti al di stri buti on i s 1/ and i ts
variance i s 1/
2
. (Remember that the time between arrivals is exponentially distributed and the number
of arrivals per unit of time is Poi sson distributed.)
3. Ot her Arri val Charact eri st i cs
a. Arri val pat t erns. The arri vals at a system are f ar more control l abl e that is general . The si mpl est of all arri val -control
devi ces i s the posti ng of busi ness hours. Some servi ce demands are cl earl y uncontrol l abl e, such as emergency media
demand on a ci tys hospi tal are control l abl e to some extent by, say, keepi ng ambul ance drivers i n the servi ce regi on
i nf ormed of the status of thei r respecti ve host hospi tal .
b. Si ze of arri val . A si ngl e arri val may be thought of as a one uni t. (A uni t i s the small est number handl ed.) A batch
arri val i s some mul ti pl e of the uni t, such as a bl ock of 1,000 shares on the PSE, a case of eggs at the processi ng plants,
or a f ami l y party at a restaurant.
c. Degree of pat i ence. A pati ent arrival i s one who wai ts as l ong as necessary unti l the servi ce f aci li ty is ready to serve
hi m or her. (Even i f arri val s grumbl e and behave i mpati entl y, the f act that they wai t i s suf f i ci ent to label them as
pati ent arri val s f or the purpose of wai ti ng li ne theory.) These are two cl asses of i mpati ent arrival s. Member of the f irst
cl ass arrive, survey both the servi ce f aci li ty and the l ength of the li ne, and then deci de to l eave. Those i n the second
cl ass arrive, vi ew the si tuati on, j oi n the wai ti ng li ne, and then, af ter some peri od of ti me, depart. The behavi or of the
f i rst type i s termed bal k i ng, whil e the second i s termed r enegi ng.
n! i s def i ned as n(n-1)(n-2)(2)(1)
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 72
The Servi ci ng Syst em
The Queui ng System consi sts pri mari l y of the wai ti ng li ne(s) and the avail abl e number of servers. The f ol l owi ng discusses
i ssues pertai ni ng to wai ti ng li ne characteri sti cs and management, l i ne structure, and servi ce rate.
1. The Wai t ing Li ne. Factors to be consi der wi th wai ti ng li nes i nclude the l i ne l ength, number of li nes, and queue
di sci pli ne.
a. Lengt h. I n practi cal sense, an i nf i ni te l i ne si mpl y one that i s very l ong i n terms of the capaci ty of the servi ce system.
Exampl es i n i nfi ni t e pot ent i al l engt h are a l i ne of vehi cl es backed up f or mil es at a bri dge crossi ng and customer who
must f orm a l i ne around the bl ock as they wai t to purchase ti ckets at a theater. Gas stati on, l oadi ng docks, and parki ng
l ots have l i mi t ed li ne capaci t y caused by l egal restri cti on or physi cal space characteri sti cs.
b. Number of Lines. A si ngl e li ne or si ngl e f i l e is, of course, one l i ne onl y. The term mult i pl e l i ne ref ers to the si ngl e l i nes
that f orm i n f ront of two or more servers or to si ngl e li nes that converge at some central redi stri buti on poi nt.
c. Queue Di sci pl i ne. A queue disci pli ne i s a pri ori ty rul e or set of rul es f or determi ni ng the order of servi ce to
customers i n a wai ti ng li ne. The rul es sel ected can have a dramati c ef f ect on the systems overal l perf ormance. The
number of customer i n li ne, the average wai ti ng ti me, the range of vari abil i ty i n wai ti ng l i ne, and the ef f i ci ency of the
servi ce f acil i ty are just a f ew of the f actors af f ected by the choi ce of pri ori ty rul es. (F i r st come; fi r st ser ve (F CF S), shor t est
pr ocessi ng ti me, r eser vati on fi r st , emer genci es fi r st , hi ghest -pr ofi t customer fi r st , l argest or der s fi r st , best cust omer fi r st , l ongest wai ti ng ti me i n
l i ne, seni ori t y or di sabl e lane, and soonest pr omi sed dat e are some exampl es of pri ori ty rul es.)
2. Servi ce Ti me Di st ri but i on. Another i mportant f eature of the wai ti ng structure i s the ti me the customer or uni t spends
wi th the server once the servi ce has started. Wai ti ng li ne f ormul as general l y speci f y ser vi ce r at e as the capaci ty of the server
i n number of uni ts per ti me peri od (such as 12 compl eti ons per hour) and not as servi ce ti me, whi ch might average f i ve
mi nutes each. A const ant servi ce ti me rul es states that each servi ce takes exactl y the same ti me. As i n constant arri vals, thi s
characteri sti c i s generall y l i mi ted to machi ne-cont rol l ed operati ons. When servi ce ti mes are random, a good approxi mati on
of them can be given by the exponenti al di stri buti on. When usi ng the exponenti al distri buti on as an approxi mati on of the
servi ce ti mes, we wi ll ref er to as the average number of uni ts or customers that can be served per ti me peri od.
3. Li ne St ruct ure. The f l ow of i tems to be servi ced may go through a si ngl e li ne, mul ti pl e l i nes, or some mi xtures of the
two. The choi ce of f ormat depends partl y on the vol ume of customers served and partl y on the restri cti ons i mposed by
sequenti al requi rements governi ng the order i n whi ch servi ce must be perf ormed.
a. Si ngl e channel , si ngl e phase e.g., one-person barbershop
b. Si ngl e channel , mul t i pl e phase e.g., car wash (vacuumi ng, wetting, washi ng, rinsi ng, dryi ng, window
cl eaning, and parki ng)
c. M ul t i pl e channel , si ngl e phase e.g., bank tel l ers wi ndows and checkout counters i n high-volume
department stores
d. M ul t i pl e channel , mul t i pl e phase e.g., admission of hospi tal pati ents
e. M i xed. Under this general heading we consi der two categori es (1) mul ti pl e-to-si ngl e channel structures
(e.g., toll gate in NLEX, bri dge crossing, or subassembl y l ines) and (2) al ternate path structure.
The condi t i on of t he cust omer exi t i ng t he syst em (back t o source popul at i on or not ?)
Once a customer i s served, two exi t f ates are possi bl e: (1) the customer may return to the source popul ati on and immedi atel y
become a competi ti ve candi date f or servi ce agai n: or (2) there may be a l ow probabi l i ty of reservi ce. The f i rst case can be
i ll ustrated by a machi ne that has been routi nel y repai red and returned to duty but may break down agai n: the second can be
i ll ustrated by a machi ne that has been overhaul ed or modi f i ed and has a l ow probabi l i ty of reservi ce over the near f uture. I n a
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 73
l i ghter vei n, we might ref er to the f i rst as the recurri ng-common-col d case and to the second as the appendectomy-onl y-
one case.
Waiting Line Method
M easures of Wai t i ng Line Performance
They relate to potential customer dissatisfaction and costs.
1. The average number of customer waiting, either in line or in the system.
2. The average time customer waits, either in line or in the system.
3. System utilization, which refers to the percentage of capacity utilized.
4. The implied cost of a given level of capacity and its related waiting line.
5. The probability that an arrival will have to wait for service.
I nfi ni t e M odel
Al l assume a Poisson arrival rate. Moreover, the model pertai ns to a system operati ng under st eady-stat e condi ti ons; that i s, they
assume the average arri val and servi ce rate are stabl e. The f our model are descri bed bel ow
Prope rtie s of Some Sp e c ific Wa iting Line M od e ls
M odel Layout Service Phase
Source
Population
Arrival
Pattern
Queue
D isci pline
Service
Pattern
Permissi ble
Queue
Length
T ypi cal Example
1
2
3
4
Si ngl e channel
Si ngl e channel
Mul t i pl e channel
Si ngl e channel
Si ngl e
Si ngl e
Si ngl e
Si ngl e
I nf i ni te
I nf i ni te
I nf i ni te
Fi ni t e
Poi sson
Poi sson
Poi sson
Poi sson
FCFS
FCFS
FCFS
FCFS
Exponent i al
Const ant
Exponent i al
Exponent i al
Unl i mi t ed
Unl i mi t ed
Unl i mi t ed
Unl i mi t ed
D ri ve-i n t el l er at bank;
one-l ane t ol l bri dge:
Rol l er coast er ri des i n
amusement part
Part s count er i n aut o
agency
Machi ne breakdown and
repai r i n a f act ory
Infinite-Source Symbols
Symbols Represents Symbols Represents
L
q
L
s
r
Customer arrival rate
Service rate per server
The average number of customer waiting for
service
The average number of customer in the
system(waiting and/or being served)
The average number of customer being serve
The system utilization
W
q
W
s
1/
P
0
P
n
M
L
MAX
The average time customer wait in line
The average time customers spend in the system
(waiting in line and service time)
Service time
The probability of zero units in the system
The probability of n units in the system
The number of servers (channels)
The maximum expected number waiting in line
Ba sic Re la tionships
1. Syst em Ut i l i zat ion: Thi s ref ers to the rati o of demand (as measured by the arri val rate) to suppl y or capaci ty (as
measured by the product of the number of servers, M , and the servi ce rate, ).
=
M
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 74
2. The average number of customer bei ng served:
r =
{ N ote: and must be i n the same uni ts (e.g., customers per hour, customer per mi nute)}
Accordi ng to Li ttl es Law, f or stabl e system, the average number of customer i n li ne or i n the system i s equal to the average
customer arri val rate mul ti pli ed by the average ti me i n li ne or i n system. That i s,
L
s
= W
s
and L
q
= W
q
3. The average number of customer
a. Wai ti ng i n li ne f or servi ce: L
q
(Model dependent. Obtai n usi ng a tabl e or f ormul a.)
b. I n the system (l i ne pl us bei ng served): L
s
= L
q
+ r
4. The average ti me customers are
a. Wai ti ng i n li ne:
W
q
=
L
q
b. I n the system:
W
s
= W
q
+
1
=
L
s
Sample Problem Sets for Waiting Line Management
1. Customers arri ve at a bakery at an average rate of 16 per hour on weekday morni ngs. The arri val distri buti on can be
descri bed by a Poi sson di stri buti on wi th a mean of 16. Each cl erk can serve a customer i n an average of three mi nutes;
thi s ti me can be descri bed by an exponenti al di stri buti on wi th a mean of 3.0 mi nutes.
a. What are the arrival and servi ce rate?
b. Compute the average number of customer bei ng served at anyti me.
c. Supposed i t has been determi ned that the average number of customer wai ti ng i n l i ne i s 3.2. Compute the average
number of customer i n the system (i .e., wai ti ng i n li ne or bei ng served), the average ti me customers wai t i n l i ne, and
the average ti me i n the system.
d. D etermi ne t he system uti li zati on f or M = 1, 2, and 3 servers.
2. An ai rli ne is planni ng to open satel li te ti cket desk i n a new shoppi ng pl aza, staf f ed by one ti cket agent. I t is esti mated that
request f or ti ckets and i nf ormati on wil l average 15 per hour, and request wi ll have a Poisson distri buti on. Servi ce time i s
assumed to be exponenti al l y distri buted. Previ ous experi ence wi th si mi lar satel li te operati on suggests that mean servi ce
ti me shoul d be average about three mi nutes per request. D etermi ne each of the f ol l owi ng.
a. System uti li zati on.
b. Percentage of ti me the server (agent) wil l be i dl e.
c. The expected number of customer wai ti ng to be served.
d. The average ti me customer wi ll spend i n the system.
e. The probabi l i ty of zero customers i n the system and the probabi l i ty of f our customers i n the system.
A
n
s
w
e
r
s
:
1
.
a
)
T
h
e
a
r
r
i
v
a
l
r
a
t
e
i
s
g
i
v
e
n
i
n
t
h
e
p
r
o
b
l
e
m
:
=
1
8
c
u
s
t
o
m
e
r
s
p
e
r
h
o
u
r
.
T
h
e
s
e
r
v
i
c
e
t
i
m
e
c
a
n
b
e
c
h
a
n
g
e
d
t
o
a
c
o
m
p
a
r
a
b
l
e
h
o
u
r
l
y
r
a
t
e
b
y
f
i
r
s
t
r
e
s
t
a
t
i
n
g
t
h
e
t
i
m
e
i
n
h
o
u
r
s
a
n
d
t
h
e
n
t
a
k
i
n
g
i
t
s
r
e
c
i
p
r
o
c
a
l
.
T
h
u
s
,
(
4
m
i
n
u
t
e
s
p
e
r
c
u
s
t
o
m
e
r
)
/
(
6
0
m
i
n
u
t
e
s
p
e
r
h
o
u
r
)
=
1
/
1
5
=
1
/
.
I
t
s
r
e
c
i
p
r
o
c
a
l
i
s
=
1
5
c
u
s
t
o
m
e
r
s
p
e
r
h
o
u
r
;
b
)
r
=
1
.
2
c
u
s
t
o
m
e
r
s
;
c
)
L
s
=
4
.
8
c
u
s
t
o
m
e
r
s
,
W
q
=
0
.
2
0
h
o
u
r
s
p
e
r
c
u
s
t
o
m
e
r
,
o
r
1
2
m
i
n
u
t
e
s
,
W
s
0
.
2
6
7
h
o
u
r
s
o
r
a
p
p
r
o
x
i
m
a
t
e
l
y
1
6
m
i
n
u
t
e
s
;
d
)
M
=
2
p
=
.
6
0
,
M
=
3
p
=
.
4
0
M
=
4
p
=
.
3
0
2
.
a
)
p
=
.
7
5
;
b
)
2
5
p
e
r
c
e
n
t
c
)
L
q
=
2
.
2
5
c
u
s
t
o
m
e
r
s
d
)
W
s
0
.
2
0
h
o
u
r
s
o
r
1
2
m
i
n
u
t
e
s
e
)
P
0
=
.
2
5
P
4
=
.
0
7
9
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 75
Name: Score:
Course & Year: Schedule: Date:
Problem Set for Waiting Line Management
1 . Noel O . Pa n Jr. Lub e Inc. op era t es a f a st lub e a nd oil cha ng e g a ra g e. O n a t y p ica l d a y , cust omers arr ive at
t he ra t e of t hr ee p er hour a nd lub e job s a r e p erf ormed a t a n a ver ag e ra t e of one ever y 15 minut es. The
mecha nics op era t e a s a t ea m on one car at a t ime. Assuming Poisson a r r iva ls a nd ex p onent ia l ser vice, f ind .
a . Ut iliza t ion of t he lub e t ea m.
b . The a vera g e numb er of ca rs in line.
c. The a vera g e t ime a ca r wa it s b ef or e it is lub ed .
d . The t ota l t ime it t a kes t o g o t houg h t he syst em (t hat is, wa it ing in line p lus lub e t ime).
e. The a vera g e numb er of ca rs in t he syst em.
f . The a vera g e t ime a ca r is lub ed .
g . The p r ob a b ilit y of no car t o b e lub ed a nd oil cha ng ed in t he g a ra g e.
h. The p r ob a b ilit y t hat f ive ca r s b eing lub ed and oil cha ng ed in the ga r ag e.
2 . Ba r na chea Bus Liner is p lanning t o op en sa t ellit e t icket d esk in a Shoe Ma rt Cit y - Ba g uio, sta f f ed b y one
t icket ag ent . It is est ima t ed tha t r eq uest f or t ickets a nd inf or mat ion will a ver ag e 15 p er hour , a nd r equest
will ha ve a Poisson d ist r ib ut ion. Ser vice time is assumed t o b e ex p onent ia lly d ist r ibut ed. Pr evious ex p er ience
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 76
wit h simila r sat ellit e op era t ion sugg est s t hat mea n ser vice t ime should b e a ver a g e a b out t hr ee minut es p er
r eq uest . Det er mine ea ch of t he f ollowing .
a . Sy st em ut iliza t ion.
b . Per cent ag e of t ime t he ser ver (a g ent ) will b e id le.
c. The ex p ect ed numb er of cust omer wa it ing t o b e served .
d . The a vera g e t ime cust omer will sp end in t he sy st em.
e. The a vera g e numb er of cust omer b eing ser ved .
f . The ex p ect ed t ime a cust omer will wa it in t he line.
g . The p r ob a b ilit y of zer o cust omers in t he syst em.
h. The p r ob a b ilit y of f our cust omers in t he sy st em.
3 . Ba r ney Ta x i a nd Ha uling Ser vice p lans t o ha ve ca bs a t Vict or y Liner st a t ion. The exp ect ed a rr iva l r at e is 4 .8
cust omer s p er hour , and t he ser vice ra t e (includ ing ret ur n t ime t o t he st at ion) is exp ect ed t o b e 1 .5 p er hour .
How ma ny ca b s will b e need ed t o a chieve a n avera g e time in line of 20 minut es or less?
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 77
G. OPERATING & CONTROLLING THE P/O SYSTEM - Part II: PROJECT MANAGEMENT
Project: a unique, one time operational activity or effort.
Project Planning
The general management process is concerned with the planning,
organization, and control of an ongoing process or activity such as the
production of a product or delivery of a service. Project Management is
different in that it requires a commitment of resources and people to an
important undertaking that is not repetitive and involves a relatively short period
of time, after which the management effort is dissolved. A project has a unique
purpose, it is temporary, and it draws resources from various areas in the
organization; as a result, it s subject to more uncertainty than the normal
management process. Thus, the features and characteristics of the project
management process tend to be unique.
Project management encompasses three other major processesplanning,
scheduling, and control. It also includes a number of the more prominent
elements of these processes.
1. Planning ( Project Team, Work Breakdown Structure, Responsibility
Assignment Matrix)
2. Scheduling (Gantt Chart, CPM/PERT)
3. Control (Stay on Schedule, Risk Analysis and Quality Control, Cost
Control)
Elements of a Project Plan
Project plans generally include the following basic elements.
Objectives a detailed statement of what the project is to accomplish and
how it will achieve the companys goal and meet the strategic plan; and
an estimate of when it needs to be completed, the cost and the return.
Project scope a discussion of how to approach the project, the
technological and resource feasibility, the major task involved, and a
preliminary schedule; includes a justification of the project and what
constitutes project success.
Contract requirements a general structure of managerial, reporting, and
performance responsibilities, including a detailed list of staff, suppliers,
subcontractors, managerial requirements and agreements, reporting
requirements, and a projected organizational structure.
Schedules a list of all major events, tasks, and subschedules, from
which a master schedule is developed.
Resources the overall project budget of all resources requirements and
procedures for budgetary control.
Personnel identification and recruitment of personnel required for the
project team, including special skills and training.
Control procedure for monitoring and evaluating progress and
performance including schedules and cost.
Risk and problem analysis anticipating and assessing uncertainties,
problems, and potential difficulties that might increase the risk of project
delays and/or failure and threaten project success.
Project teams are made up of individuals from various areas and departments
within a company.
Matrix organization: a team structure with members from functional areas,
depending on the skills required.
Project manager is often under great pressure (time and budgetary constraints)
Scope statement: a document that provides an understanding, justification, and
expected results of a project.
Statement of work: a written description of the objective of a project (use for
determining work performance and for bidding of the subcontractors and
suppliers)
Work breakdown structure (WBS): breaks down a project into components,
subcomponents, activities, and tasks.
Organizational Breakdown Structure (OBS): a chart that shows which
organizational units are responsible for work items.
Responsibility Assignment Matrix (RAM): shows who is responsible for the
work in a project.
WBS Activities Hardware/Installation
OBS Units
1.1.1
Area
Prep
1.1.2
Tech/
Engineer
1.1.3
Wiring
1.1.4
Connections
Hardware engineering
Systems engineering
Software engineering
Technical support
Electrical staff
Hardware vendor
Quality manager
Customer/supplier
liaison
3
1
2
3
1
3
3
2
3
1
2
3
1
3
2
2
3
3
3
Level of responsibility: 1 = overall responsibility, 2 = performance responsibility, and 3 =
support
For global projects to successful, cultural differences, idiosyncrasies, and
issues must be considered as important parts of the planning process.
Project Scheduling
Gantt chart: a graph or bar chart with a bar for each project activity that shows
the passage of time. It has been a popular project scheduling tool since its
inception and is still widely used today. It is the direct precursor of the
CPM/PERT technique.
Month
0 2 4 6 8 10
Activity
1.Design house and obtain financing
2..Lay foundation
3.Order and receive materials
4..Build house
5.Select paint
6.Select carpet
7.Finish work
1 3 5 7 9
Month
Order System
Personnel Software Hardware
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 78
Slack: the amount of time an activity can be delayed without delaying the project.
Project Control
Project control is the process of making sure the project progresses toward a
successful completion. It requires that the project be monitored and progress be
measured so that any deviations from the project plan, and particularly the project
schedule, are minimized. If the project is found to be deviating form the planthat
is, it is not on schedule, cost overruns are occurring, activity results are not as
expected, and so onthen corrective actions must be taken. The following are the
key elements of project control:
1. Time Management: the process of making sure the project schedule does not
slip and it is on time. This requires monitoring of individual activity schedules
and frequent updates. If the schedule is being delayed on an extent that
jeopardizes the project success, then the project manager may have to shift
resources to accelerate critical activities.
Time-cost tradeoff: activities may have slack time, and resources can be
shifted from them to activities that are not on schedule, however, this can also
push the project cost above budget.
2. Cost Management: is often closely tied to time management because of the
time cost tradeoff occurrences. If the schedule is delayed, costs tend to
increase in order to get the project back on schedule.
3. Quality Management: quality management and control are an integral part of
the project management process. The process requires that project work be
monitored for quality and that improvements be made as the project
progresses just the same as in a normal production or manufacturing
operation. Poor-quality work increases the risk of project failure, just as a
defective part can result in a defective final product if not corrected.
4. Performance Management: the process of monitoring a project and
developing timed (i.e., daily, weekly, monthly) status reports to make sure that
goals are being met and the plan is being followed. It compares planed target
dates for events, milestones, and work completion with dates actually
achieved to determine whether the project is on schedule or behind schedule.
Key measures of performance include deviation from the schedule, resource
usage, and cost overruns.
Earned value analysis (EVA) is a standard procedure for numerically
measuring a projects progress, forecasting its completion date and cost and
measuring schedule and budget variation. Example given, schedule
variance compares the work performed during a time period with the work
that was schedule to be performed, and cost variance is the budgeted cost
of work performance minus the actual cost of the work.
5. Communication: communication needs for project and program management
control in todays global business environment tend to be substantial and
complex. The distribution of design documents, budget and cost documents,
plans, status reports, schedules, and schedule changes in a timely manner is
often critical to project success.
6. Enterprise Project Management: refers to the management and control of a
companywide portfolio of projects. In the enterprise approach to managing
projects, a companys goals are achieved through the coordination of
simultaneous projects. The company grows, changes, and adds value by
systematically implementing projects of all types across the enterprise.
CPM/PERT
Critical Path Method (CPM), developed in 1956 by research team at E.I. du Pont de
Nemours & Company, Inc., to initiate a project to developed a computerized
system to improve the planning, scheduling, and reporting of the companys
engineering programs (including plant maintenance and construction projects)
Program Evaluation and Review Technique (PERT) developed by research team
from Navy Special Project Office, and consulting firm developed a similar
network approach for the design of management control system for the
development of the Polaris Missile Project (a ballistic missile-firing nuclear
submarine).
Both CPM and PERT are derivatives of the Gantt chart, and are very similar.
However these are the differences:
1. With CPM a single estimate for activity was used that did not allow for any
variation in activity timesactivity times were treated as if they were known
for certain, or deterministic. With PERT, multiple time estimates were used
for each activity that allowed for variation in activity timesactivity times were
treated as probabilistic.
2. In PERT, activities were represented as arcs, or arrowed lines, between two
nodes or circles, whereas in CPM activities were represented as the nodes or
circles.
CPM/PERT uses a network to depict the precedence relationships among activities.
Project Network
1. Activity-on-node (AON): nodes represent activities, and arrows show
precedence relationships.
C
C C C C
C
Dummy activity is inserted into the network to show a precedence relationship, but it
does not represent any actual passage of time
2. Activity-on-arrow (AOA): arrows represent activities and nodes are events for
points in time.
The Critical Path
A network path is a sequence of connected activities that runs from the start to the end of
the network. The network in preceding illustration has several paths through it. In fact,
close observations of this network show for paths, identified as A, B, C, and D:
Design house
and obtain
financing
Lay
foundation
Order and
rec eive
materials
Build
house
Finish
work
Dummy
Selec t
paint
Selec t
c arpet
3
3
2
1
3
1
1
1
0
Lay
foundation
Order
materials
Order
materials
Lay
foundation
Dummy
3
3
Correc t Wrong
2 4 2
1
2 4
3 5 6
7
Design house
and obtain
financing
Lay
foundation
Build house
Order and
rec eive
materials
Selec t
paint
Selec t
c arpet
Finish
work
Ac tivity
number
Ac tivity
time
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 79
A:
B:
C:
D:
1-2-4-7
1-2-5-6-7
1-3-4-7
1-3-5-6-7
Critical path is the longest path through a network; it is the minimum project completion
time.
Path A:
Path B:
Path C:
Path D:
1-2-4-7
3 + 2 + 3 + 1 = 9 months
1-2-5-6-7
3 + 2 + 1 + 1 + 1 = 8 months
1-3-4-7
3 + 1 + 3 + 1 = 8 months
1-3-5-6-7
3 + 1 + 1 + 1 + 1 = 7 months
Conditions:
1. Activity 3 will start at 3 months
2. Activity 4 will start at 5 months
3. Activity 6 will start at 6 months
4. Activity 7 will finish at 9 months
Activity Scheduling
Activity
number
(AN)
Earliest
start
(ES)
Earliest
finish
(EF)
1
0 3
3 0 3
Activity
duration
(AD)
Latest
start
(LS)
Latest
finish
Node configurations
1. Earliest start time (ES): the earliest time an activity can start.
Earliest start time = maximum (Earliest finish time of immediate predecessors)
Earliest finish time = ES + t
2. Forward pass: starts at the beginning of a CPM/PERT network to determine the
earliest activity times.
3. Earliest finish time (EF): is the earliest time plus the activity time.
4. Latest start time (LS): the latest time an activity can start without delaying critical
path time.
Latest start time = LF + t
Earliest finish time = minimum (Latest start time of immediate following activities
5. Latest finish time (LF): the latest time an activity can be completed and still
maintain the project critical path time.
6. Backward pass: determines latest activity times by starting at the end of a
CPM/PERT network and working forward.
Activity Slack
Notice that for the activities on the critical path, the earliest times and latest times are
equal. This means that these activities on the critical path must start exactly on time and
cannot be delayed at all. If the start of any activity on the critical path is delayed, then the
overall project time will be increased. The activities on the critical path can be determined
by seeing for which activities ES=LS or EF= LF. For activities not on the critical path for
which the earliest and latest start times (or earliest and latest finish times), are not equal,
slack time exists.
Slack is the amount of time an activity can be delayed without affecting the overall
project duration. In effect, it is extra time available for completing an activity.
Slack (S) is computed using either of the following formulas: S = LS ES or S = LF
EF. For example, the slack for activity 3 is S = LS ES = 4 3 = 1 month. If the start of
activity 3 were delayed for one month, the activity could still be completed by month 5
without delaying the project completion time.
Activity LS ES LF EF
Slack S
*1
*2
3
*4
5
6
*7
* = critical path
0
3
4
5
6
7
8
0
3
3
5
5
6
8
3
5
5
8
7
8
9
3
5
4
8
6
7
9
0
0
1
0
1
1
0
It shows there is no slack for the activities on the critical path (marked with an
asterisk); activities not on the critical path have slack.
Shared slack means that the sequence of activities 3-5-6 can be delayed two months
jointly without delaying the project, but not three months.
Slack is beneficial to the project manager because it enables resource to be
temporarily diverted from activities with slack and used for other activities that might be
delayed for various reasons or for which the time estimate has proved to be inaccurate..
The times for the network activities are simply estimates, for which there is usually not
a lot of historical basis (since project tends to be unique undertakings). As such, activity
time estimates are subject to quite a bit of uncertainty. However, the uncertainty inherent
in activity time estimates can be reflected to a certain extent by using probabilistic time
estimates instead of the single, deterministic estimates.
Probabilistic Activity Times
It is rare that activity time estimates can be made with certainty. Project activities are
likely to be unique with little historical evidence that can be used as a basis to predict
activity times. Probabilistic time estimates reflect uncertainty of activity times.
In the PERT-type approach to estimate activity times, three time estimates for each
activity are determined, which enables us to estimate the mean and variance of a beta
distribution of the activity times.
Beta distribution: a probability distribution traditionally used in CPM/PERT. We
assume that the activity times can be described by a beta distribution for several
reasons:
1. The beta distribution mean and variance can be approximated with three time
estimates.
2. The beta distribution is continuous, but has not predetermined shape (such as the bell
shape of the normal curve). It will take on the shape indicatedthat is, be skewed
by the time estimates given.
The three times estimates for each activity are the most likely time (m), the optimistic
time (a), and the pessimistic time (b). The most likely time is a subjective estimate of the
activity time that would most frequently occur if the activity were repeated many times.
The optimistic time is the shorter possible time to complete the activity if everything went
right. The pessimistic time is the longest possible time to complete the activity assuming
everything went wrong. The person most familiar with an activity or the project manager
makes these subjective estimates to the best of his or her knowledge and ability.
These three times estimates are used to estimate the mean and variance of a beta
distribution, as follows.
Mean (expected time) =
a + 4m + b
6
Variance:
2
=
b a
6
where:
a =
m =
b =
optimistic time
most likely time
pessimistic time
2
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 80
Example of a project network with probabilistic time estimates:
Solution
As an example of the computation of the individual activity mean times and variance,
consider activity 1. The three time estimates (a = 6, m = 8, b = 10) are substituted in the
formula as follows:
t =
a + 4m + b
=
6 + 4(8) + 10
= 8 weeks
6 6
2
=
b - a
=
10 - 6
= 4/9 week
6 6
Activity Time Estimates for the example above:
Time Estimates (weeks)
Mean
time
Variance
Activity a m b t
2
1
2
3
4
5
6
7
8
9
10
11
6
3
1
2
2
3
2
3
2
1
1
8
6
3
4
3
4
2
7
4
4
10
10
9
5
12
4
5
2
11
6
7
13
8
6
3
5
3
4
2
7
4
4
9
0.44
1.00
0.44
2.78
0.11
0.11
0.00
1.78
0.44
1.00
4.00
Activity Earliest and Late Times and Slack
Activity t
2
ES EF LS LF S
1
2
3
4
5
6
7
8
9
10
11
8
6
3
5
3
4
2
7
4
4
9
0.44
1.00
0.44
2.78
0.11
0.11
0.00
1.78
0.44
1.00
4.00
0
0
0
8
6
3
3
9
9
13
16
8
6
3
13
9
7
5
16
13
17
25
1
0
2
16
6
5
14
9
12
21
16
9
6
5
21
9
9
16
16
16
25
25
1
0
2
8
0
2
11
0
3
8
0
Critical Path = 2-5-8-11 (no available slack times)
To determine the project variance, we sum the variance for the activities on the critical
path.
2
=
2
2 +
2
5 +
2
8 +
2
11
= 1.00 + 0.11 + 1.78 + 4.00
= 6.89 weeks
Probabilistic Network Analysis
The CPM/PERT method assumes that the activity times are statistically independent,
which allows us to sum the individual expected activity times and variance to get an
expected project time and variance. It is further assumed that the network mean and
variance are normally distributed. This assumption is based on the central limit theorem
of probability, which for CPM/PERT analysis and our purpose states that if the number of
activities is large enough and the activities are statistically independent, then the sum of
the means of the activities along the critical path will approach the mean of a normal
distribution.
Probabilistic analysis of a CPM/PERT network is the determination of the probability
that the project will be completed within a certain period given the mean and variance of
a monthly distributed project completion time. The value Z is computed using the
following formula:
Z =
x
where:
=
x =
=
Z =
variance
tp = project time
The proposed project time
Number of standard deviation x is from the mean
Normal Distribution of Project Time
= tp x
Project Crashing and Time-Cost Trade-Off
The project manager is frequently confronted with having to reduce the scheduled
completion time of a project to meet a deadline. In other words, the manager must finish
the project sooner than indicated by the CPM/PERT network analysis. Project duration
can often be reduced by assigning more labor to project activities, in the form of
overtime, and by assigning bore resources (material, equipment, and so on). However
additional labor and resources increase the project cost. Thus, the decision to reduce the
project duration must be base on an analysis of the tradeoff between time and cost.
Project crashing is a method for shortening the project duration by reducing the time of
one (or more) of the critical project activities to less than its normal activity time. This is
reduction in the normal activity time is referred to as crashing. Crashing is achieved by
devoting more resources, usually measured in terms of dollars, to the activities to be
crashed.
Crashing; reducing project time by expending additional resources.
Project crashing
Objective: the objective of project crashing is to reduce project duration while minimizing
the cost of crashing. Since the project completion can be shorten only by crashing
activities on the critical path, it may turn out that not all activities. However, as activities
are crashed. The critical path may change, requiring crashing of previous noncritical
activities to reduce the project completion time even further.
Normal activity cost: cost required to complete the activity
Crash time: an amount of time an activity is reduced.
Crash cost: is the cost of reducing activity time.
Activity 1 can be crashed a total of five weeks (normal time crash time = 12 7 = 5
weeks) at a total crash cost of $2000 (crash cost normal cost = $5000 - $3000 =
$2000). Dividing the total crash cost by the total allowable crash time yields the crash
cost of per week. Total crash cost / total crash time = $2000/5weeks = $400 per week.
1
2 2
Probabilit y
Z
2
3
4
5
6
7
8
9
10
11
S F
6,8,10
3,6,9
1,3,5
2,4,12
2,3,4
3,4,5
2,2,2
3,7,11
2,4,6
1,4,7
1,10,13
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 81
PROBLEM SET FOR PROJECT MANAGEMENT
Name: Date:
Course & Year: Schedule: Score:
The School of Information and Computing Science at Saint Louis University has outgrown its offices in
the Diego Silang Building (S) and is moving to Maryheights Campus Bakakeng (D), which has more space.
The move will take place during the three-week break between end of summer semester and the beginning of
the first semester. Movers will be hired from the universitys physical plant to move the furniture, boxes of
books, and files that the faculty will pack. The school has hired a local retail computer firm to move its office
computers so they will not be damaged. Following is a list of activities, their precedence relationships, and
probabilistic time estimates for this project:
Activity Description
Activity
Predecessor a m b
Time Estimate
(days)
a
b
c
d
e
f
g
h
i
j
Pack D offices
Network D offices
Pack S offices
Movers move D offices
Paint and clean D offices
Move computers
Movers move S offices
Computer installation
Faculty move and unpack
Faculty set up computers and offices
-
-
-
a
d
b, e
b, c, e
f
g
h, i
1
2
2
1
2
1
3
2
3
1
3
3
4
3
5
2
6
4
4
2
5
5
7
4
8
2
8
5
6
4
___
___
___
___
___
___
___
___
___
___
1. Calculate the most probable or mean time. Indicate the answer on the table above.
2. Construct the Network Diagram in the space provided below. Avoid DUMMIES, and ERASURES.
Determine the early start and the early finish of each activity. What is earliest finish of the project?
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 82
3. Reconstruct the Network Diagram in the space provided below. Determine the late start and the late finish
of each activity. What is the latest start of the project?
4. Tally the results following the table below.
Activity
Early Start
(ES)
Early Finish
(EF)
Late Start
(LS)
Late Finish
(LF)
Slack
(LS ES) or
(LF EF)
Is it the
Critical Path?
CRITICAL PATH:
5. Compute for the expected project completion.
6. Determine the variance of the activity.
7. Determine the probability of completing the project in 18 days or less.
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 83
i . OPERATI NG & CONTROLLI NG THE P/ O SYSTEM Par t I : I NVENTORY M ANAGEM ENT
Lear ni ng Obj ect i ves: Af t er compl et i ng t hi s modul e, st udent shoul d be abl e t o:
1. D ef i ne i nvent or y, i nvent or y management , quant i t y di scount ;
2. I dent i f y t ypes of i nvent or y;
3. Li st some r easons w hy or gani zat i ons have i nvent or y;
4. D i scuss t he concer ns and obj ect i ve of i nvent or y management ;
5. Li st t he t w o basi c i nvent or y deci si ons;
6. Li st r equi r ement s f or ef f ect i ve i nvent or y management ;
7. D i scuss t he r el evant i nvent or y cost s;
8. D i scuss t he t w o gener al t ypes of i nvent or y syst ems
9. Cl assi f y i nvent or y syst ems
10. D o comput at i ons r el at ed t o i nvent or y management and quant i t y di scount
I nvent or y r ef er s t o st ock or stor e of goods, may r el at e t o w or k- i n- pr ocess goods
Types of I nvent or i es
1. Raw mat er i al s and pur chased par t s
2. W or k - i n- pr ocess or par t i al l y compl et ed goods
3. Fi ni shed goods ( manuf act ur i ng f i r ms) or mer chandi se ( r et ai l st or es)
4. Repl acement par t s, t ool s, accessor i es, and suppl i es
5. Goods- i n- t r ansi t t o w ar ehouse or end user ( cust omer s)
W hy Or gani zat i on has I nvent or y
1. To meet ant i ci pat ed ..
st ocks ar e hel d t o sat i sf y ex pect ed aver age demand
2. To pr ot ect agai nst ..
st ocks, w hi ch ar e st ock s i n ex cess of aver age demand t o compensat e f or var i abi l i t y i n demand
and l ead t i me, can r educe t he r i sk of shor t ages.
3. To smoot h .. r equi r ement s
i nvent or i es ar e st or ed dur i ng .. per i ods t o meet over l y hi gh r equi r ement s dur i ng cer t ai n
seasonal per i ods.
4. To . component s of t he pr oduct i on t o di st r i but i on syst em
.. i nvent or i es ar e st ock ed bet w een successi ve oper at i ons t o mai nt ai n cont i nui t y of succeedi ng
oper at i ons i n case of . i n pr ecedi ng oper at i ons
Raw mat er i al s and suppl i es i nven t or i es ser ve as . bet w een t he suppl i er and t he f i r m
W or k - i n- pr ocess i nvent or i es ser ve as buf f er bet w een successi ve w or kst at i ons
Fi ni shed goods i nvent or i es ser ve as buf f er bet w een t he f i r m and i t s cust omer r equi r ement s
5. To per mi t
.. i nvent or i es ex i st t hr ough t he pr oduct i on t o di st r i but i on syst em due t o t he f act t hat pr oduct i on
oper at i ons t i me ( i .e., t hey ar e not i nst ant aneous) .
6. As a r esul t of t ak i ng advant age of .
To mi ni mi ze pur chasi ng and cost s, a f i r m may buy i n quant i t i es t hat ex ceed i mmedi at e
r equi r ement s. Thi s necessi t at es st or i ng t he ex cess f or l at er use.
The Concer ns of I nvent or y M anagement
1. . of t he st ock w hen needed t o max i mi ze l evel of cust omer ser vi ce
2. . of or der i ng and car r yi ng t he i nvent or y t o max i mi ze i nvent or y cost s
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 84
The Obj ect i ve of I nvent or y M anagement
To achi eve sat i sf act or y l evel of cust omer ser vi ce w hi l e k eepi ng i nvent or y cost s w i t hi n r easonabl e bounds: i .e.,
t r y t o achi eve a .. bet w een t he concer ns of i nvent or y management
The Basi c I nvent or y M anagement D eci si ons
1. .. .. ( si ze) How much t o or der
2. .. .. ( t i mi ng) W hen t o or der
Requi r ement s f or Ef f ect i ve I nvent or y M anagement
1. A syst em to .. . of t he i nvent or y on hand and on or der .
2. A r el i abl e . .. t hat i ncl udes an i ndi cat i on of possi bl e f or ecast er r or .
3. Know l edge of . . and l ead t i me var i abi l i t y.
4. Reasonabl e est i mat es of i nvent or y .. , .. ...., and ..
5. A f or pr i or i t i zi ng i nven t or y i t ems.
The Rel evant I nvent or y M anagement Cost s
1. Cost of t he i t em or . cost s
The sum pai d to t he suppl i er f or t he i t em r ecei ved
The di r ect manuf act ur i ng cost
.. i f t he i t em uni t cost i s const ant f or al l quant i t i es or der ed
.. i f t he i t em uni t cost var i es w i t h quant i t y or der ed, e.g., quant i t y di scount
2. Or der i ng or set up or cost s
The cost s i ncur r ed by pl aci ng an or der f or and r ecei vi ng t he i t em or i ncur r ed as set up cost s w hen i tems i s
manuf act ur ed
I ncl ude cost of post age, phone cal l s t o vendor s, l abor cost i n pur chasi ng and account i ng, r ecei vi ng cost s,
r ecor ds k eepi ng and pur chase or der suppl i es
3. Car r yi ng or hol di ng or . cost s
The cost s associ at ed w i t h havi ng t he i nvent or y on hand
I ncl ude i nsur ance, w ar ehouse r ent al s, heat i ng and l i ght i ng, t ax es, l osses due t o pi l f er ages, spoi l ages, or
br eak ages
. cost f or havi ng capi t al t i ed up i n i nvent or y
4. St ockout or .. cost s
The cost s associ at ed w i t h demand w hen st ocks have depl et ed
Loss of sal es
Cl assi f i cat i on Syst em
An i mpor t ant aspect of i nvent or y management i s t hat i t ems hel d i n i nvent or y ar e not of equal
i mpor t ance i n t er ms of pesos i nvest ed, pr of i t pot ent i al , sal es or usage vol ume, or st ock- out
penal t i es. I t w oul d be unr eal i st i c t o devot e equal at t ent i on t o each of t hese i t ems. A mor e
r easonabl e appr oach w oul d be t o al l ocat e cont r ol ef f or t s accor di ng t o t he r el ati ve i mpor t ance
of t he var i ous i t ems i n i nvent or y.
The A- B- C Appr oach cl assi f i es i nvent or y i t ems accor di ng t o some measur e of i mpor t ance,
usual l y annual peso usage ( i .e., peso val ue per uni t mul t i pl i ed by annual usage r at e) and t hen
al l ocat es cont r ol ef f or t s accor di ngl y. Thr ee cl asses of i t ems ar e used:
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 85
A ( ver y i mpor t ant ) gener al l y account about 15% to 20% of t he number of i t ems i n i nvent or y but about 60% t o 70% of
t he peso usage. A i t ems shoul d r ecei ve cl ose at t ent i on t hr ough f r equent r evi ew of amount on hand and cont r ol over
w i t hdr aw al s, w her e possi bl e, t o mak e sur e t hat cust omer ser vi ce l evel s ar e at t ai ned.
B ( moder at el y i mpor t ant )
C ( l east i mpor t ant ) may account f or about 60% of t he number of i t ems but onl y about 10% of t he peso usage as an
i nvent or y. C i t ems shoul d r ecei ve onl y l oose cont r ol ( t w o- bi n syst em, bul k or der s)
The A- B- C concept can be appl i ed as a gui de to cycl e count i ng, w hi ch i s a physi cal count f i t ems i n i nvent or y. The
pur pose of cycl e count i ng i s t o r educe di scr epanci es bet w een t he amount s i ndi cat ed by i nvent or y r ecor ds and t he
act ual quant i t i es of i nvent or y on hand. Accur acy i s i mpor t ant because i naccur at e r ecor ds can l ead t o di sr upt i ons i n
pr oduct i on, poor cust omer ser vi ce, and unnecessar i l y hi gh i nven t or y car r yi ng cost s.
The k ey concer ns of cycl e count i ng f or management ar e:
1. How much accur acy i s needed?
2. W hen shoul d cycl e count i ng be per f or med?
3. W ho shoul d do i t ?
PERPETUAL SYSTEM ver sus PERI ODI C SYSTEM
Featur e Per petual System Per i odi c System
1 . Recor d k eepi n g
Per pet u al ; cou n t ed each t i me an
addi t i on or w i t h dr aw al i s made
Per i odi c; cou n t ed on l y at r evi ew per i od;
i n vol ves l oose mon i t or i n g of i n ven t or y
2 . Resou r ces r equ i r ed to
mai n t ai n
M or e, si n ce per pet u al r ecor d k eepi n g
3 . Si ze of i n ven t or y
Less t h at of t h e per i odi c syst em
M or e t h an t hat of t h e per pet u al syst em,
si n ce saf et y st ock i s added to avoi d
st ock ou t s i n bet w een r evi ew per i ods
4 . Type of i n ven t or y syst em
For h i gh er - pr i ced, cr i t i cal or
i mpor tan t i t ems
I nvent or y Syst em
1. Per pet ual Syst em
I nvent or y M odel I nvent or y D eci si on For mul a
Fi x ed- Or der Quant i t y or
Economi c Or der Quant i t y ( EOQ)
model
How much t o or der ?
EOQ i s const ant ; t he same
quant i t y or der ed each t i me
2 2 DCo DCo
EOQ
Ch iC
W hen t o or der ?
Or der i s pl aced w hen t he quant i t y
on hand dr ops to t he r eor der l evel
( R)
R dL
2. Per i odi c Syst em
I nvent or y M odel I nvent or y D eci si on For mul a
Fi x ed- Ti me Per i od or Fi x ed- Or der
I nt er val model
How much t o or der ?
q i s var i abl e; t he or der quant i t y
var i es each t i me an or der i s pl aced
( ) q d T L SS I
SS z T L
W hen t o or der ?
Or der i s pl aced w hen t he r evi ew
per i od ( T) ar r i ves
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 86
Assumpt i on of t he Basi c Economi c Or der Quant i t y ( EOQ) M odel
1. Ther e i s onl y one i t em i nvol ved
2. Annual demand ( usage) i s k now n
3. D emand ( usage) r at e i s uni f or m
4. The i t em uni t cost does not var y w i t h t he or der si ze ( no quant i t y di scount s)
5. Each or der i s r ecei ved i n a si ngl e del i ver y
6. Lead t i me i s k now and does not var y
7. Or der i ng or set up cost i s t he same r egar dl ess of t he amount or der ed
8. I nvent or y hol di ng cost i s based on aver age i nvent or y l evel
9. Al l demands f or t he i t em w i l l be sat i sf i ed ( no st ock out , no backor der )
Symbol s used i n t he Basi c EOQ M odel
TC =
D =
C =
R =
Tot al annual i nvent or y cost
Annual demand, uni t s
Cost per uni t of t he i t em, P/ uni t
Quant i t y t o be or der ed ( t he opt i mum
amount i s t he economi c or der
quant i t y, EOQ)
Reor der poi nt , uni t s
Co =
Ch =
L =
Set up or or der i ng cost , P/ or der
Annual hol di ng or car r yi ng cost per
uni t of aver age, P/ uni t ( of t en hol di ng
cost i s ex pr essed as per cent age of t he
i t em uni t cost , such t hat Ch = i C,
w her e i i s t he per cent car r yi ng cost )
Lead t i me, days
Quant i t y D i scount
Quant i t y di scount s ar e t he pr i ce r educt i on f or l ar ge or der s of f er ed t o cust omer s to i nduce t hem t o buy i n l ar ge
quant i t i es.
Ex . Pr i ce l i st f or a ex t r a- w i de gauze st r i ps of Bayombong Sur gi cal Suppl y Company
Or der Quant i t y Pr i ce per box
1 t o 44
45 t o 69
70 or mor e
P 80
68
56
2
Q D
TC Ch Co
Q
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 87
The cust omer must w ei gh t he pot ent i al benef i t s of r educed pur chase pr i ce and f ew er or der s t hat w i l l r esul t f r om
buyi ng i n l ar ge quant i t i es agai nst t he i ncr ease i n car r yi ng cost s caused by hi gher aver age i nvent or i es. The buyer s goal
w i t h quant i t y di scount s i s to sel ect t he or der quant i t y t hat w i l l mi ni mi ze t ot al cost s ( i .e., sum of car r yi ng cost, or der i ng
cost , and pur chasi ng cost ) .
Tot al cost = Car r yi ng cost + Or der i ng cost + Pur chasi ng cost
2
Q D
TC Ch Co PD
Q
P = u n i t pr i ce
Sampl e Pr obl em
The mai nt enance depar t ment of a l ar ge hospi t al uses about 816 cases of l i qui d cl eanser annual l y. Or der i ng cost s ar e
P480, car r yi ng cost ar e P160 per case a year , and t he new pr i ce schedul e i ndi cat es t hat or der s of l ess t han 50 cases w i l l
cost P800 per case, 50 t o 79 cases w i l l cost P720 per case, 80 t o 99 cases w i l l cost P680 per case, and l ar ger or der s w i l l
cost P640 per case. D et er mi ne t he opt i mal or der quant i t y and t he t ot al cost .
Sol ut i on:
D = 816 cases per year Co = P480 Ch = P160 per case per year
Range Pr i ce
< 50
50 t o 79
80 t o 99
100 >
P 800
720
680
640
Economi c Or der Quant i t y = _
2Co
Ch
= _
2( 816) ( 480)
160
= 70 cases
The 70 cases can be bought at P720 per case si nce 70 f al l s i n t he r ange of 50 t o 79 cases. The t otal cost t o pur chase 816
cases per year , at t he r at e of 70 cases per or der w i l l be:
TC = Car r yi ng cost + Or der i ng cost + Pur chasi ng cost
TC7 0 =
2
Ch +
Co + PD =
70
2
160 +
816
70
480 + 720( 81 6) = P598, 716
Si nce l ow er cost r anges ex i st , each must be check ed agai nst t he mi ni mum cost gener at ed by 70 cases at P720 each. I n
or der t o buy at P680 per case, at l east 80 cases must be pur chased. The t ot al cost at 80 cases w i l l be;
TC8 0 =
2
Ch +
Co + PD =
80
2
160 +
816
80
480 + 680( 81 6) = P566, 176
To obt ai n a cost of P640 per case, at l east 100 cases per or der ar e r equi r ed. The t otal cost w i l l be;
TC1 0 0 =
2
Ch +
Co + PD =
100
2
160 +
816
100
480 + 640( 816) = P534, 157
Ther ef or e, si nce 100 cases per or der yi el ds t he l ow est t ot al cost , 100 cases i s t he over al l opt i mal or der quant i t y.
Ex er ci se
Benguet El ect r i c Company uses 4,000 t oggl e sw i t ches a year . Sw i t ches ar e pr i ced as f ol l ow s: 1 t o 499, P36 each; 500 t o
999, P34 each; and 1,000 or mor e, P32.80 each. I t cost s appr ox i mat el y P720 t o pr epar e an or der and r ecei ve i t , and
t he car r yi ng cost s ar e 18% of pur chase pr i ce per uni t on an annual basi s. D et er mi ne t he opt i mal or der quant i t y and t he
t ot al annual cost .
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 88
Si ngl e- Per i od I ndependen t D emand I nven tor y M odel
W h en manager s ar e f aced w i t h on e- t i me i n ven t or y buys, t he so- cal l ed "New sboy M odel " can be an i mpor tan t t ool f or an al ysi s. Th e
n ame comes f r om t h e cl assi cal pr obl em f aced by n ew spaper ven dor s on a dai l y basi s. Th ese ven dor s mu st pl ace a on e- ti me or der
f or t h e n ex t day's paper s w i t h no ch ance of r eor der i n g sett i n g u p t h e pr esses agai n w ou l d be cost pr oh i bi t i ve. Si n ce deman d f or
t h e n ex t day i s u n k n ow n , t h e n ew spaper ven dor may or der "t oo mu ch" or "not en ou gh". I f t h i s ven dor or der s t oo mu ch, t h en s/ h e
w i l l f ace t h e cost of bei n g over st ock ed. Li k ew i se, i f t h e vendor or der s t oo f ew paper s, t h en s/ h e w i l l i n cu r t h e cost of bei n g
u n der st ock ed. Th e ven dor i s i n t er est ed i n bal an ci n g t h ese t w o cost s so t h at ex pect ed pr of i t i s max i mi zed.
Th e t oo- much cost , or t h e cost of bei n g over st ock ed, Cos, i n cl u des t h e cost of t h e l ef t over pr odu ct n et of an y sal vage val ue. Th i s
sal vage val u e mi gh t be n egat i ve, t h ough , i n t h e case of h avi n g t o di spose of ex cess, u sel ess pr oduct . I f t h e n ew sboy must t ak e h i s
l ef t over paper s t o t h e dump, t h er e i s a cost of tak i n g t h em th er e. Th e n ot - enou gh cost , or t h e cost of bei n g u n der stock ed, Cu s,
i n cl u des t h e l ost pr of i t f r om sal es n ot made an d t h e cost of th e l oss of cu stomer goodw i l l . Al t h ou gh t h e amou n t of goodw i l l l ost i s
di f f i cu l t t o est i mat e, i t n eeds t o be accou n t ed f or i n some man n er .
Th er e ar e many si mi l ar si t u at i on s u n der w h i ch manager s f ace t h i s same k i n d of di l emma. W h en a speci al i zed pi ece of equ i pmen t
i s pu r ch ased, a on e- t i me spar e par t s or der may be pl aced at t h e same t i me. Th e l ocal gr ocer y st or e bu ys br ead on ce a day an d
many ot h er per i sh abl e pr odu ct s on a r egu l ar basi s. Ai r l i n es mu st deci de how many r eser vat i on s t o t ak e f or a gi ven f l i gh t t h ey
may over book si n ce t h ey don 't k n ow h ow man y peopl e w i l l act u al l y sh ow u p and u se t h ei r r eser vat i on .
M ar gi nal Anal ysi s Appr oach
A mar gi n al anal ysi s appr oach compar es th e mar gi n al pr of i t (M P) of sel l i ng one mor e u n i t w i t h t he mar gi n al l oss (M L) associ at ed w i t h t hat
addi t i on al u n i t of i n ven t or y. I f P i s t h e pr obabi l i t y t h at on e mor e u ni t i s sol d, t hen t h e deci si on r ul e i s
I f ( P) M P ( 1 - P) M L, t h en st ock on e mor e u n i t .
W i t h some al gebr a,
( P) M P M L - ( P) M L
( P) M P + P( M L) M L
P ( M P + M L) M L
P M L / ( M P + M L)
An Ex ampl e
A h ar dw ar e st or e makes a on ce- a- year bu y of l i ve Ch r i st mas t r ees. Th e t r ees ar e pu r ch ased at a
cost of $10 each an d sel l f or $3 8 each . M an agement of t he st or e est i mat es deman d t o aver age 4 00
u n i t s and t o var y bet w een 30 0 and 5 0 0 t r ees, w i t h an y amou nt i n t h at r an ge equ al l y l i kel y. The
deman d di st r i bu t i on i s show n i n t he adj acent ex h i bi t . For ever y t r ee l ef t over at t he en d of t he season,
t h e st or e pays a $1 di sposal f ee, w h i l e f or ever y t r ee shor t , i t est i mat es t hat i t l oses $2 i n l ost cust omer
goodw i l l .
To f i nd t he best quan t i t y t o st ock , t he equat i on f or P i s used.
P M L / ( M P + M L)
Th e mar gi n al l oss, M L (or t h e cost of over st ock i n g, or t he cost of havi ng t oo mu ch ) i s t he cost of buyi n g t he t r ee pl us t h e di sposal cost . The
mar gi n al pr of i t , M P ( or t he cost of un der st ock i n g, or t he cost of n ot havi n g enou gh ) i s t he cost of l ost goodw i l l an d l ost pr of i t . Th er ef or e,
M L = $1 0 + $1 = $1 1,
M P = $2 8 + $2 = $3 0 , an d t h us,
P 1 1 / [ 30 + 1 1 ] = 0.2 7.
P= 0 .2 7 i s t he pr obabi l i t y t h at one mor e u ni t w i l l sel l , or t h e r i gh t - h and t ai l of t he di st r i bu t i on.
Th u s, t he pr obabi l i t y of meet i n g demand i s 1 - P, or 0.7 3, t h e l ef t - h and t ai l of t he di st r i but i on . For
t h e u n i f or m di st r i but i on , t he t ar get i n ven t or y l evel i s 0 .73 ( 5 0 0- 3 0 0 ) + 3 00 = 4 4 6 , as show n i n
t h e adj acent ex hi bi t . I n t h e l angu age of t he cur r en t si t u at i on , t h e har dw ar e st or e w an t s t o cover
7 3 % of t he possi bl e demand si t uat i ons. Not i ce t h at t he cost of bei n g un der st ocked, t he M P, i s
si gn i f i cant l y h i gher t h an t he cost of bei ng over st ock ed, t he M L. Th er ef or e, t he pr ef er en ce w oul d
be f or over st ock i n g, or st ock i n g mor e t h an t he ex pect ed demand of 4 0 0 u n i t s.
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 89
FIXED ORDER QUANTITY MODEL
1. Given the following data: annual demand, D = 1,000 units; ordering cost, S = P200 per order; holding cost,
H = P50 per unit per year; lead time, L = 5 days; and cost per unit, C = P500.
a. Find the economic order quantity.
b. Find the reorder point.
c. What would be your ordering policy for this item?
d. Find the total annual inventory cost,
2. A fastfood outlet uses 120 eight-ounce plastic cups each day and plans to be open 360 days a year. The cups
cost P9per dozen; Ordering costs are P200 order; and carrying costs are 50 percent of the item unit cost
(since space is a premium).
a. Find the economic order quantity, if delivery is instantaneous.
b. Currently, cups are ordered every 30 days. 1) Compare the current ordering quantity against the optimal
order quantity; and 2) Compare the current against the optimal total annual inventory cost. Which
ordering policy is more advantageous in terms of cost savings? By how much?
3. A toy manufacturer uses approximately 32,000 silicon chips annually. The chips are used at a steady rate
during the 240 days the plant operates. Annual holding cost is P27 per chip and ordering cost is P1, 080.
a. Find the economic order quantity.
b. Find the reorder point.
c. What would be your ordering policy for this item?
d. Find the total annual cost of ordering and carrying silicon chips.
4. Each year, Barney Company purchases 20,000 units of an item that costs P640 per unit. The cost of placing
an order is P480, and the cost to hold the item in inventory for one year is P50.
a. Calculate the economic order quantity.
b. What is the average inventory level, assuming that the minimum inventory level is zero?
c. Determine the total annual ordering cost and the total annual holding cost for the item if the economic
order quantity is used.
5. A large bakery buys sugar in 50-kilograms bags. The bakery uses an average of 1,344 bags a year. Preparing
an order and receiving a shipment of sugar involves a cost of P135. Annual carrying costs are P630 per bag.
The bakery operates 280 days per year.
a. Determine the economic order quantity.
b. What is the average number of bags on hand?
c. When should the bakery order for more sugar?
d. How many times per year will the bakery order for sugar?
FIXED TIME PERIOD MODEL
1. A drug store orders its antibiotics every two weeks when salesperson visits from one of the pharmaceutical
companies. Amoxicillin is one of its most prescribed antibiotics, with an average daily demand of 1,500
capsules. The standard deviation of daily demand was derived by examining prescriptions filled over the last
three months and was found to be 500 capsules. It takes nine days for the order to arrive. The drug store
PRODUCTION &/ or OPERATIONS MANAGEMENT Page 90
manager has set a service level of 99 percent. The sales person has just arrived, and there are currently
25,000 capsules in stock.
a. How much additional stock must be carried to minimize the risk of stock out?
b. How many capsules of Amoxicillin should be ordered if lot size per order is 100 capsules?
2. The weekly demand for a product is 770 units. The standard deviation of daily demand is 10 units. The firm
reviews its inventory every 28 days. Each order for the product arrives 9 days after placement. At the time
of review, there are 10 units backorder. If 95 percent of all demand is to be satisfied from items in stock.
a. How many units should be the safety stock?
b. What is the products inventory status?
c. How many units should be ordered for this period?
QUANTITY DISCOUNT
1. Benguet Electric Cooperative, Inc. uses 4,000 toggle switches a year. Switches are priced as follows: 1 to
499, P36 each; 500 to 999, P34; and 1,000 or more, P32.80 each. It costs approximately P720 to prepare an
order and receive it, and the carrying costs are 18% of purchase price per unit on an annual basis. Determine
the optimal order quantity and the total annual cost.
2. A small manufacturing firm uses roughly 3,400 pounds of chemical dye a year. Currently the firm purchases
300 pounds per order and pays P150 per pound. The supplier has just announced that orders of 1,000 pounds
or more will be filled at a price of P100 per pound. The manufacturing firm incurs a cost of P5, 000 each
time it submits an order and assigns an annual holding cost of 17 percent of the purchase price per pound.
a. Calculate the order size that will minimize the total cost.
b. If the supplier offered the discount at 1,500 pounds instead of 1,000 pounds, what order size would
minimize total cost?
SINGLE-PERIOD INVENTORY MODEL
1. The local supermarket buys lettuce each day to ensure really fresh produce. Each morning any lettuce that is
left from the previous day is sold to a dealer that resells it to farmers who use it to feed their animals. This
week the supermarket can buy fresh lettuce for Php200 a box. The lettuce is sold for Php500 a box. Past
history says that tomorrows demand for lettuce average 250 boxes with a standard deviation of 34 boxes.
How many boxes of lettuce should the supermarket purchase tomorrow?
2. Semper Fidelis Pizza orders all of its pepperoni, olives, anchovies, and mozzarella cheese to be shipped
directly from Italy. An American distributor stops by every four weeks to take orders. Because the orders
are shipped directly from Italy, they take three weeks to arrive.
Semper Fidelis Pizza orders an average of 150 pounds of pepperoni each week, with a standard
deviation of 30 pounds. Semper Fidelis prides itself on offering only the best-quality ingredient and a high
level of service, so it wants to ensure a 98 percent probability of not stocking out on pepperoni.
Assume that the sales representative just walked in the door and there are currently 500 pounds of
pepperoni in the walk-in cooler. How many pounds of pepperoni would you order?
PREPARED & ADAPTED: napoleon i barnachea jr second semester ay 2011-2012 october 26, 2011 18:13:37