NETWORKTECHNI QUES(PERT&CPM)

KHARAGPURCOLLEGE
BCA4TH SEMI
STER
PROGRAM EVALUATIONANDREVI
EW TECHNI
QUE(
PERT)
7.
1:AnInt
roduct
ion

Apr ojectdefinesacombinationofinter
rel
atedacti
vi
ti
est hatmustbeex ecutedin
a cer t
ain orderbeforethe entir
et askcan becompl et ed.Theact i
vit
iesare
i
nterrelatedinal ogi
csequencesi nt hesensethatsomeact i
vit
iescannotstart
untilothersar ecompleted.Anact iv
ityinapr oj
ectisusual l
yv i
ewedasaj ob
requiri
ngt i
meandr esourcesforit
scompl eti
on.Ingener al,aprojectisaone–
ti
meef fort
;thatis,t
hesamesequenceofact i
vit
iesmaynotber epeatedinthe
future.

Int hepast, t
heschedul i
ngofapr oj
ect( overtime)wasdonewi thl i
tt
lespeci fi
es
thest artandf i
nishf oreachact i
v i
tyonahor i
zontalti
mescale.I tsdisadv antage
i
st hatt heinterdependencybet weent hedifferentacti
vi
ti
es(whichmai nlycont rol
s
thepr ogressoft hepr oj
ect)cannotbedet erminedf r
om thebarchar t.
Thegr owing
'
compl ex i
ti
es oft odays pr ojecthav e demanded mor e systemat i
c and mor e
effect i
v eplanningt echniqueswi tht heobj ectiveofoptimizi
ngt heef fi
ciencyof
execut i
ngt hepr oject.Effi
ciencyher ei mpliesef f
ecti
ngal mostreduct i
oni nt he
ti
me r equir
ed t o compl etet he pr ojectwhi l
e accounti
ng fort he economi c
feasi bil
ityofusingav ail
abler esources.
Proj ectmanagementhasev olvedasanew f i
eldwiththedev el
opmentoft wo
anal ytictechniquesf orplanning, schedul i
ng,andcont r
oll
ingofprojects.

7.2 Backgr ound:

Thesear et hecriticalpat hmet hod( CPM)andthepr ojectev aluat
ionandr eview
technique(PERT) .Thet wot echniquesweredevelopedbyt wodi f
ferentgroups
almostsi multaneousl y( 1956-1958)CPM wasf i
rstdev elopedbyE. IduPoutde
Nemour sandcompanyasanappl i
cat
iontoconst
ruct i
onpr ojectsandwasl ater
extendedt oamor eadv ancedst atusbyMauchlyassoci ates.PERT, ontheot her
handwasdev elopedf ort heU. S.Navybyaconsul tingf ir
mf orscheduli
ngt he
researchanddev elopmentact i
viti
esforthePol
ari
smi ssileprogram.

1
PERTandCPM ar ebasi cal lyt i
me–or ientedmet hodsi nt hesenset hatt heybot h
l
eadt ot hedet er minat ionofat imeschedul e.Al thought het womet hodswer e
dev elopedi ndependent ly,t heyar est ri
kingl ysi milar .Per hapst hemosti mportant
differencei st hator iginallyt het imeest imat esf ort heact i
vitieswer eassumed
det ermi nisti
ci nCPM andpr obabi l
ist i
cinPERT.Today,PERTandCPM act ual l
y
compr ise one t echni que and t he di fferences ,i fany ,ar e onl y hi stori
cal.
Consequent l
y ,bot ht echni ques wi llbe r ef err
ed t o as " pr ojectschedul ing"
techni ques.
Proj ectschedul ingbyPERT- CPM consi stoft hreebasi cphases
1.Pl anni ng
2.Schedul i
ngand
3.Cont rolli
ng.
7.3 Pl anni ngPhase
Thepl anni ngphasei si niti
at edbybr eakingdownt hepr ojecti ntodi stinctactivi
ties
.Thet i
meest i
mat esf ort heseact iviti
esar et hendet er minedandanet work( or
arrow) di agram i sconst ruct ed wi th each ofi t
s ar cs( arr
ows)r epr esent i
ng an
act i
v i
ty.The ent ire ar row di agram gi ves a gr aphi cr epr esent at i
on oft he
i
nt erdependenci esbet weent heact iv i
ti
esoft hepr oject .Theconst ruct i
onoft he
arrow di agram asapl anni ngphasehast headv ant ageofst udy ingt hedi f
ferent
j
obsi ndet ail.per hapssuggest ingi mpr ov ementbef or et hepr ojecti sactual l
y
execut ed.Mor eimpor tantwi l
lbei tsuset odev el
opaschedul ef ort hepr oj
ect.
7.4 Schedul i
ngphase
Theul ti
mat eobj ectiveoft heSchedul ingphasei st oconst ructat imechar t
showi ngt hest artandf inisht i
mesf oreachact i
vityaswel lasi tsr elat i
onshipt o
otheract i
vit
iesi nt hepr oject .Inaddi t
ion,t heSchedul emustpi npoi ntt hecrit
ical
(i
nv i
ew oft ime)act ivi
tiest hatr equi respeci alat tent ioni ft hepr ojecti stobe
compl etedont i
me.Fort henoncr iti
calact iviti
est heschedul emustshow t he
amountofsl ackorf l
oatt i
met hatcanbeusedadv ant ageousl ywhensuch
act i
v i
tiesar edel ay edorwhenl i
mi tedr esour cesar et obeusedef f
ect ively.
7.5 Cont rolli
ng
Thef i
nalphasei npr ojectmanagementi spr ojectcont rol.Thi si ncludest heuseof
thear rowdi agram andt het imechar tformaki ngper i
odi cpr ogr essr eports.The
net wor kmayt husbeupdat edandanal yzedand,i fnecessar y,anewschedul ei s
det ermi nedf ort her emai ningpor tionoft hepr oject .

2
7.6 Pr oj ectManagement
Projectmanagementi sconcer nedwi ththeov eral
lpl anni ngandco- ordinat ionof
a pr oject f rom concept ion t o compl eti
on ai med at meet i
ng t he st ated
requirement sandensur ingcompl etionont ime, withincostandt orequi redqual it
y
standar ds.Pr oj ectmanagementi snor mal l
yr eser vedf orf ocused,non-r epet i
ti
v e,
ti
me- l
imi tedact i
v i
tieswi t
hsomedegr eeofr i
skandt hatar ebey ondt heusual
scopeofoper ationalact iv
itiesf orwhi cht heor gani zat i
oni sr esponsi ble.Apr oject
i
sat empor aryendeav ouri nvolvingaconnect edsequenceofact ivit
iesanda
rangeofr esour ces,whi chi sdesi gnedt oachi ev easpeci f
icanduni queout come
andwhi choper at eswi t
hint ime,costandqual i
tyconst raintsandwhi chi sof ten
usedt oi ntroducechange.
7.7 Char act er i
st i
cofapr oject
 Auni que, one- timeoper at i
onal activit
yoref fort
 Requi rest hecompl etionofal argenumberofi nt errelatedact ivi
ties
 Est abl i
shedt oachi ev especi fi
cobj ect i
ve
 Resour ces, suchast imeand/ ormoney , arelimi t
ed
 Ty pical lyhasi tsownmanagementst ruct ure
 Needl eader ship
 Theappl icat i
onofacol lectionoft oolsandt echni quest odi r
ectt heuseof
diver ser esour cest owar dst heaccompl ishmentofauni que,compl ex,one
ti
met askwi thintime, costandqual i
tyconst raints.
 Usedt het echni quesofoper ational r
esear cht opl ant heopt i
mum useof
resour ces.
 Oneoft heset echni queswast heuseofnet wor kst or epr esentasy stem
ofr elat edact i
viti
es
7.8 Pr oj ectManagementPr ocess
 Pr oj ectpl anni ng
 Pr oj ectschedul i
ng
 Pr oj ectcont rol
 Pr oj ectt eam - madeupofi ndiv i
dual sfrom v ar i
ousar easanddepar tment s
withi na company
 Mat ri
xor gani zati
on- at eam st ructurewi t hmember sf rom f unctional ar eas,
dependi ngonski ll
sr equi red

3
  Project Manager - most i mpor tant member of pr oj
ect team Scope
statement- adocumentt hatpr ov i
desanunder standing,justi
fication,and
ex pectedresultofapr oject
 Statementofwor k- writ
tendescr ipti
onofobj ect
iv esofapr oject
 Or ganizati
onal Br eakdown St ruct ur
e - a char tt hat shows whi ch
organi zati
onalunitsarer esponsi bleforworki t
ems
 Responsi bili
tyAssignmentMat r i
x- showswhoi sr esponsibl
ef orwor ki na
pr
oject
7.
9 Wor kbreakdownst ructur e
 A met hod of br eaki
ng down a pr oj
ect i nto individual el ement s
(component s,subcomponent s,act i
vi
ties and t asks)i n a hierar chical
structurewhi chcanbeschedul edandcost
 Itdef inest askst hatcan becompl eted i
ndependent l
yofot hert asks,
facili
tati
ng r esource al locat i
on, assi gnment of r esponsi bil
it
ies and
measur ementandcont rol ofthepr oject
 Itisfoundat i
onofpr oj
ectpl anni ng
 Iti sdev el
opedbef oreident ifi
cat i
onofdependenci esandest imat ionof
act i
vi
tydur ati
ons
 Itcanbeusedt oidentit
yt het asksi ntheCPM andPERT

7.
10 Pr ojectPlanni ng
• ResourceAv ailabili
tyand/ orLi mit
s
– Duedat e, latepenal ti
es,earlycompl eti
onincent
ives
– Budget
• Activi
tyInformat i
on
– I dentifyal lrequiredact i
viti
es
– Est imat et her esour cesrequired( ti
me)tocompl eteeachacti
vit
y
– I mmedi ate pr edecessor(s) to each act ivi
ty needed t o
createi nter r
elationships
7.
11 Pr ojectSchedul ingandCont rolTechni ques
1.Crit
icalPathMet hod( CPM)
2.Pr
ogram Ev aluat i
onandRev i
ewTechni que( PERT)
7.
12 Pr ojectNet wor k
• Networ kanal ysisi sthegener alnamegi vent ocert
ainspecif
ictechniques
whichcanbeusedf ort heplanning, managementandcont rolofpr
ojects

4
• Useofnodesandar rows
Arrows An ar r
ow l eads f r
om t ailto head di r
ecti
onal lyIndicate
ACTIVITY,at i
meconsumi ngef fortthatisr equir
edt operform apar toft he
work.
Nodes Anodei sr epresent edbyaci rcl
eIndicateEVENT, apoi ntin
ti
mewher eoneormor eact i
viti
esst artand/orfi
nish
• Act iv
ity
– At askoracer tainamountofwor krequir
edint heproject
– Requi rest i
met ocompl ete
– Repr esent edbyanar r
ow
• DummyAct i
vity
– I ndicatesonl yprecedencer elati
onships
– Doesnotr equi
reanyt imeofef f
ort
• Ev ent
– Si gnal sthebegi nningorendi ngofanact i
vit
y
– Desi gnat esapoi ntint i
me
– Repr esent edbyaci rcl
e( node)
• Net work
– Showst hesequent ialrel
at i
onshipsamongact i
vit
iesusi ngnodes
andar rows
 Act iv
ity
-on- node( AON)
nodesr epr esentactivit
ies,andar r
owsshowpr ecedencer elati
onships
 Act iv
ity
-on- arrow( AOA)
arr
owsr epr esentactivi
tiesandnodesar eeventsforpointsint i
me

Lay 3 Dummy
foundat
ion
2 0 Buil
d Fi
nis
3 1 hous h
1 2 4 e 6 work 7
Design Orderand
house 3 1 1
receiv
e Sel
ec Select
and mat er
ial t 1 carpe
obtain s pai
nt t
5

Fi
g7.
1:AOAPr
ojectNet
wor
kforHouse

5
 Lay Bui
ld
f
oundat
ions house

2 4 Fi
nishwor
k
2 3
7
St
art 1 1
3
Design 6
house 3
and 1 5 1
obtai
n 1 Sel
ect
Orderand car
pet
r
ecei
vemat er
ial
s Sel
ect
pai
nt

Fi
g7.
2:AONPr
ojectNet
wor
kforHouseSi
tuat
ions
i
nnet
wor
kdi
agram

B
A

C
Fi
g7.
3: Amustf
ini
shbef
oreei
therBorCcanst
art

A
C
B

Fi
g7.
4:bot
hAandBmustf
ini
shbef
oreCcanst
art

6
 A C
B D

Fi
g7.
5:bot
hAandCmustf
ini
shbef
oreei
therofBorDcanst
art

A B

Dummy

C
D
Fi
g7.
6:Amustf
ini
shbef
oreBcanstart
bot
hAandCmustfini
shbefor
eDcanst
art

7
 Lay 3
Lay
Dumm
foundat
io
2 0
2 3
1
Or
der 2 4
Or
der

(
a) Incorr
ect (
b) Corr
ect
precedence pr
ecedence

Fi
g7.
7:Concur
rentAct
ivi
ti
es

7.
12 Uncer
tai
ntyi
npr
ojectschedul
ing
Duri
ngprojectexecution,howev er
,areal
-l
if
eprojectwi l
lneverexecuteexactlyas
i
twaspl annedduet ouncertainty
.Itcanbeambi gui
tyresult
ingfrom subjective
esti
matest hatarepr onet ohumaner r
orsoritcanbev ari
abili
tyari
si
ngf r
om
unexpect
edev ent
sorr i
sks.Themai nreasont hatthePr ojectEval
uati
onand
Revi
ew Techni que(PERT)maypr ovi
deinaccur
atei nformati
onaboutt hepr oject
completi
ont imei sduet othisscheduleuncertaint
y.Thisi naccur
acyisl ar ge
enoughtorendersuchest imatesasnothelpful
.

One possibi
li
tyto max imize solut
ion r obust
ness i stoi ncl
ude saf et
yi nthe
baseli
nescheduleinor dertoabsor bt heant i
cipateddi sr
upti
ons.Thi siscall
ed
proacti
vescheduli
ng.A pur e proact i
veschedul i
ngi saut opia,incor
porati
ng
safet
yinabasel i
neschedul ethatallowst ocopewi t
hev erypossibledisr
upti
on
wouldl ead t
o a baseline schedul
e wi thav eryl arge make-span.A second
approach,react
ive schedul
ing,consi sts ofdef ini
ng a proceduret or eactto
disr
upti
onsthatcannotbeabsor bedbyt hebaselineschedule.

CPM –CRI
TICALPATHMETHOD
7.13 AnI ntr
oducti
on:
In1957,Dupontdev el
opedapr oj
ectmanagementmet hoddesi
gnedtoaddress
the chal
lenge ofshutti
ng down chemicalpl
ant
sf ormai nt
enance and t
hen
rest
arti
ng theplantsoncet hemai nt
enancehad been complet
ed.Given t
he
complexit
yoft heprocess,theydevel
opedtheCri
ti
calPathMet hod(CPM)f or
managingsuchpr oj
ect
s.

8
CPM pr
ovi
dest
hef
oll
owi
ngbenef
it
s:

 Provi
desagr aphi
calviewoftheproject
 Predi
ctsthetimerequiredt
ocompl etethepr oject
.
 Showswhi chacti
vit
iesarecri
ti
cal
tomai nt
ainingtheschedul
eandwhi
ch
arenot.

CPM modelstheeventsandact
ivi
ti
esofaprojectasanetwork.Acti
vi
ti
esare
depi
ctedasnodesonthenet
workandevent
sthatsigni
fyt
hebegi
nningorendi
ng
ofact
ivi
ti
esar
edepict
edasarcsorl
inesbet
weenthenodes.

7.
14 Wor
kBr
eakdownSt
ruct
ures(
WBS)

Thedev el
opmentofapr ojectpl
ani spredi
catedonhav i
ngacl ear,anddetai
led
under st
andingofboththetasksinvolv
ed,theest i
matedlengthoft i
meeacht ask
willtake,thedependenciesbetweent hoset asks,andt hesequenceinwhi ch
thoset askshavet obeper f
ormed.Addi t
ional
ly,resourceavail
abil
it
ymustbe
determinedi nordertoassigneacht askorgr oupoft askstot heappropri
ate
wor ker.

Oneoft
hemethodsusedtodev
elopt
hel
i
stoft
asksi
stocr
eat
ewhati
sknown
asaWorkBr
eakdownStr
uct
ure(
WBS).

 Adef
ini
ti
on

AWor kBreakdownSt r
uct
ur e(WBS)i sahierarchicdecompositionorbr eakdown
ofapr ojectormajoracti
vi
tyintosuccessivelylevel
s,whereeachl ev elisafiner
breakdownoft heprecedi
ngone.I nf i
nalform aWBSi sverysi mil
arinst r
ucture
andl ayouttoadocumentout l
i
ne.Eachi t
em ataspeci fi
cl evelofaWBSi s
number edconsecuti
vel
y(e.g.10,20,30,40,50).Eachitem att
henex tlev
elis
number edwithi
nthenumberofi tsparentit
em (e.g.10.
1,10.2,10,3,10.4)

TheWBSmaybedrawnindi
agr
ammati
cfor
m(i
faut
omat
edt
ool
sar
eav
ail
abl
e)
ori
nachar
tfor
mresembl
i
nganout
li
ne.

TheWBSbegi
nswithasingl
eoveral
ltaskrepr
esent
ingthetotal
it
yofwor
ktobe
per
for
medonthepr
oject
.Thi
sbecomest henameoft hepr
ojectpl
anWBS

Usi
ngamethodologyorsystem l
i
fecy
clest
epsasaguide,theproj
ectisdivi
ded
i
ntoi
tsmaj
orsteps.Inourcasewewil
luset
heUnif
iedProcessphases.Eachof

9
thesephasesmustbebr okendownintothei
rnextlevelofdetai
l
,andeachof
those,i
ntost
il
lfi
nerl
evelofdetai
l,
unti
lamanageabletasksizei
sar r
ivedat.The
fi
rstlev
elofWor kBreakdownStruct
urefortheuni
fi
edpr ocessisil
lustr
atedin
Table6.1

Tasksateachsuccessi v
elyf i
nerlevelofdet ai
larenumberedt orefl
ectt
het ask
fr
om whi chtheywer eder i
ved.Thust hef i
rstl
eveloftaskswouldbenumber ed1.0,
2.0,3.
0,andsof orth.Eachoft hei
rsub-taskswouldhav eat wopar tnumber.The
fi
rstpartr ef
lecti
ng t hepar enttaskand t hesecond partbei ngthesub- t
ask
numberi tsel
fe.g.1. 1,1.2,1.3,etc.Aseachoft hesei si
nt ur
ndecomposedor
brokendowni ntot heircomponentt asks,thecomponent sr ecei
veanumber
comprisedofi t
spar ent’
snumberpl usauni quenumberofi t
sown.

 Anot
herDef
ini
ti
on:

Amanageabl et aski
sonewher etheexpectedresul
tscanbeeasilyidenti
fi
ed,
success,f
ail
ure,orcomplet
ionoft
hetaskcaneasilybeascer
tai
ned,thetimet o
completethetaskcaneasil
ybeesti
mated,andtheresour
cerequi
rementsoft he
taskcaneasil
ybedet er
mined.

WBS TaskDescr
ipt
ion
Number
1.
0 I
ncepti
on
1.
1 DraftProj
ectPlan
2.
0 El
aborat i
on
2.
1 PlanUserInter
views
2.
2 ScheduleUserInter
views
3.
0 Constr
uct i
on
4.
0 Transi
tion

Tabl
e7.
1Fi
rstl
evel
ofWor
kBr
eakdownSt
ruct
uref
ort
hel
i
fecy
cle

7.
15 St
epsi
nCPM Pr
ojectPl
anni
ng:

1.Specif
yt hei
ndiv
idualact
ivi
ti
es.
2.Deter
mi nethesequenceofthoseacti
vi
ti
es.
3.Drawanet workdiagr
am.
4.Esti
matet hecompleti
ontimeforeachacti
vi
ty.

1
1
 5.I
denti
fyt
hecr
it
ical
pat
h(t
helongestpatht
hroughthenet
wor
k)
6.Updat
etheCPM di
agr
am ast
heprojectpr
ogresses.

1.Speci
fyt
hei
ndi
vi
dual
act
ivi
ti
es.

From theWorkBreakdownStr
uctur
e,al
i
sti
ngcanbemadeofallt
heact
ivi
ti
esi
n
thepr oj
ect
.Thisli
stingcanbeusedast hebasi
sforaddi
ngsequenceand
durat
ioninf
ormat
ioninlat
erst
eps.

2.Det
ermi
net
hesequenceoft
hoseact
ivi
ti
es.

Someacti
vi
ti
esar edependentupont
hecompl
eti
onofot
hers.Alist
ingofthe
i
mmediat
epr edecessor
sofeachactiv
ityi
susef
ulf
orconstr
uct
ingt heCPM
net
wor
kdiagr
am.

3.Dr
awanet
wor
kdi
agr
am.

Oncetheacti
vit
iesandthei
rsequencinghavebeendefi
ned,
theCPM di
agr
am can
bedrawn.CPM or i
ginal
l
ywasdev elopedasanact i
vi
tyonnode(AON)network,
butsomeprojectpl
annerspr
efertospecif
ytheacti
vi
ti
esonthear
cs.

4.Est
imat
ethecompl
eti
ont
imef
oreachact
ivi
ty.

The ti
me requir
ed t
o compl et
e each acti
vi
ty can be est
imated usi
ng past
exper
ienceortheest
imatesofknowl edgeabl
eper sons.CPM i
sadet er
minist
ic
modelthatdoesnottakeintoaccountvari
ati
onint hecomplet
ionti
me,soonl y
onenumbercanbeusedf oranacti
vit
y’sti
meesti
mat e.

5.I
dent
if
ythecr
it
ical
pat
h

The crit
icalpat hist he l
ongest-
durat
ion pat
ht hr
ough the net
work.The
si
gnifi
canceoft hecri
ti
calpat
hi sthattheacti
v i
ti
esthatli
eoni tcannotbe
del
ayedwi thoutdel
ayi
ngtheproj
ect.Becauseofitsimpactontheenti
repr
oject
,
cri
ti
calpathanalysi
sisanimport
antaspectofpr
ojectpl
anni
ng.

Thecr i
ti
calpathcanbei
dent
if
iedbydet
ermi
ningt
hef
oll
owi
ngf
ourpar
amet
ers
foreachacti
vit
y:

 ES–earl
i
estst
artt
ime: theearl
iestt
imeatwhi
chanacti
vi
tycanbegi
n
gi
vent
hatit
spr
edecessoract
ivi
ti
esmustbecomplet
edfi
rst
.

1
1
  EF–ear liestf
inishti
me, equaltotheearl
iestst
artt
imefort
heacti
vit
yplus
theti
mer equi
redt ocompletetheactiv
ity
.
 LF– l at
estf inishtime: t hel at
estti
meatwhi chanactivi
tycanbe
completedwi t
houtdel ayi
ngt heproj
ect.
 LS– l atestst artti
me,equalt othelatestfi
nishti
memi nusthet i
me
requi
redt ocompl etetheactivi
ty.

Theslackorfloatti
mef oranacti
vit
yistheti
mebet weentheearl
iestandl
atest
startt
ime,orbetweentheearl
iestandlat
estfi
nishti
me.Slackistheamountof
ti
met hatanacti
vit
ycanbedelayedpastit
searl
iestst
artorear
li
estfi
nishwi
thout
delayi
ngtheproj
ect.

Thecr iti
calpathi st
hepaththroughtheproj
ectnetwor kinwhichnoneoft he
acti
vi
tieshav eslack,
thati
s,t
hepathforwhichLS=ESandLF=EFf oral
lacti
vi
ties
i
nt hepat h.Adel ayi
nthecri
ti
calpat
hdelaystheproject.Si
mil
arl
y,toaccel
erate
theprojectiti
snecessaryt
oreducethetot
alti
mer equiredfort
heacti
vi
tiesi
nt he
cri
ti
cal path.

6.Updat
eCPM di
agr
am

Ast heprojectpr
ogresses,
theactualt
askcomplet
ionti
meswillbeknownandt he
diagram canbeupdat edtoi ncl
udethi
sinfor
mat i
on.A new cr
it
icalpat
hmay
emer ge,and st r
ucturalchanges may be made i nthe networkifpr oj
ect
requirementschange.

7.
16 CPM l
i
mit
ati
ons

CPM was dev eloped forcomplex butf ai r

lyrouti
ne proj
ects with mini
mal
uncer
tai
ntyinprojectcomplet
ionti
mes.Forl essrouti
neprojectsther
eismor e
uncer
tai
ntyinthecompl et
ionti
mes,andt hisuncertai
ntyl
imit
st heuseful
nessof
thedetermini
sti
cCPM model .An alt
ernativeto CPM ist hePERT ( Progr
am
Eval
uati
onandRev i
ewTechnique)proj
ectpl anni
ngmodel ,whichal
lowsar ange
ofdurat
ionstobespecifi
edforeachacti
vity
.

7.17 TimeEst i
mat es
Networksprovi
deapl annedappr oacht oprojectmanagement.Tobeef fect
ive,
networksrequi
reacl eardefini
tionofal lt
hetasksthatmakeupt heprojectand
perti
nentti
meest i
mat es.Ifthepr ojectmanagercannotclari
fythenecessary
tasksandt her esourcer equir
ement s,thenno matterhow sophist
icatedthe
network,i
twil
lnotcompensat efortheseshort
comings.Anumberofclaimshav e
beenmadeaboutt hebenefit
sofpr ojectmanagementtechni
quesbutothers

1
1
havearguedthati
npart,t
hebenefi
tsar
eduetomanager
shavi
ngtoknow and
clar
if
ythetasksr
athert
hanthedi
agr
am whi
chf
oll
ows(whi
chmaybythenbesel
f
-evi
dent
).

Theobj
ecti
vesofnetworkanal
y sisar
etolocatetheacti
vi
ti
esthatmustbekeptt o
ti
me,manageactivi
ti
estomaket hemosteffect
iveuseofresourcesandl ookf or
waysofreducingthetotalpr
ojectti
me.Foranybutt hesmal l
estprojects,this
anal
ysi
sisl
ikelyt
obedoneusi ngacomput erpackage,butyourunderstandingof
theout
putwillonl
ydev el
opify ouhavesomeex per
ienceofthebasi cst epsof
anal
ysi
s.
ACTI
VITYDURATI
ONESTI
MATI
ON–BETADI
STRI
BUTI
ON

ESTI
MATEDTI
MEFORMULA

TE= A+4(
B)+C
6
WHERE:

A=MOSTOPTI
MISTI
CTI
MEB

=MOSTLI
KELYTI
ME

C=MOSTPESSI
MISTI
CTI
ME

1
1
 CONSTRUCTI
NGTHECRI
TICALPATH

I
NFORMATI
ONFORPROJECTDELTA

Act
ivi
ty Descr
ipt
ion Pr
edecessor
s Est
imat
ed
Dur
ati
on

A Cont
ractsi
gni
ng None 5

B Quest
ionnai
redesi
gn A 5

C Tar
getmar
ketI
D A 6

D Sur
veysampl
e B,
C 13

E Dev
eloppr
esent
ati
on B 6

F Anal
yzer
esul
ts D 4

G Demogr
aphi
canal
ysi
s C 9

H Pr
esent
ati
ont
ocl
i
ent E,
F,G 2

1
1
 Par
ti
alPr
ojectAct
ivi
tyNet
wor
kwi
thTaskDur
ati
ons

B E
Desig Dev
.Pr
esent
n5 6

A D F H
Cont
ra Surve Analysi Prese
ct5 y13 s4 nt2

G
C Demog.
Market 9
I
D6
RULESWHENUSI
NGTHEFORWARDPASS

1.Addal
lact
ivi
tyt
imesal
ongeachpat
haswemov
ethr
ought
he
net
wor
k(ES+Dur=EF)
,
2.Car
ryt
heEFt
imet
otheact
ivi
tynodesi
mmedi
atel
ysucceedi
ng
t
her
ecent
lycompl
etednode.ThatEFbecomest
heESoft
he
nex
tnode,
unl
esst
hesucceedi
ngnodei
samer
gepoi
nt.
3.Atamer
gepoi
nt,
thel
argestpr
ecedi
ngEFbecomest
heESf
or
t
hatnode.

17
 ACTI
VITYNETWORKWI
THFORWARDPASS

5 B 10 10 E 16
Desi
g Dev
.Pr
esent
n5 6

0 A 5 11 D 24 24 F 28 28 H 30
Contr
ac Surve Analy
si Pr
ese
t5 y13 s4 nt2

5 C 11 1 G 2
Market Demog
I
D6 .
9
 RULESFORUSI
NGTHEBACKWARDPASS
1.Subt
ractact
ivi
tyt
imesal
ongeachpat
hasy
oumov
ethr
ough
t
henet
wor
k(LF–Dur=LS)
,
2.Car
rybackt
heLSt
imet
otheact
ivi
tynodesi
mmedi
atel
y
pr
ecedi
ngt
hesuccessornode.ThatLSbecomest
heLFof
t
henex
tnode,
unl
esst
hepr
ecedi
ngnodei
sabur
stpoi
nt.
3.I
nthecaseofabur
stpoi
nt,
thesmal
l
estsucceedi
ng
LSbecomest
heLFf
ort
hatnode.

19
 ACTI
VITYNETWORKWI
THBACKWARDPASS

5 B 10 E 16
10 Dev
.Pr
esent
6 Desi
5 g 22 6 28

0 A 5 11 D 24 24 F 28 28 H 30
Cont
rac Surv
e Anal
ysi
s Pr
esent
ati
on
0 t5 5 11 y1324 24 4 28 28 2 30

11 G 20
Demogr
aph.
5 C 11
Mar
ket 19 9 28
I
D56 11
COMPLETEDACTI
VITYNETWORKWI
THCRI
TICALPATHANDACTI
VITYSLACKTI
MESI
DENTI
FIED

Cr
it
ical
Pat
hisi
ndi
cat
edi
nbol
d

5 B 10 10 E 16
1 Desi
gn 12Dev
.Pr
esent
6 5 11 22 6 28

0 A 5 11 D 24 24 F 28 28 H 30
0 Cont
ract 0 Sur
vey 0 Anal
ysi
s 0Pr
esent
ati
on
0 5 5 11 13 24 24 4 28 28 2 30

5 C 11 11 G 20
0 Mar
ket 8Demogr
aph.
I
D56 11 19 9 28
ES I
D EF

Sl
ack Task
Name

LS Dur
ati
on LF
 ACTI
VITYNETWORKDEMONSTRATI
NGLADDERI
NGTECHNI
QUE

A1 A2
Desi
gn Desi
gn A3
Desi
gn

A1 A3
Codi
ng A2 Codi
ng
Codi
ng

A2 A3
A1 Debuggi
ng Debuggi
ng
Debuggi
ng
 NETWORKDEMONSTRATI
NGHAMMOCKACTI
VITY

5 B 9 12 G 21
13 10
18 4 22 22 9 31

0 A 5 5 C 1 1 H 2 31 I 35
0 9 2 0
0 5 5 1 7 2 9 1 3 3 4 35
4 2
5 D 11 11 E 25 25 F 31
0 user 0 Codi
ng 0 Debuggi
ng
eds11 11
5 ne6 14 25 25 6 31

5 A
Hammoc
k
26
 STEPSTOREDUCETHECRI
TICALPATH
1.El
i
minat
eTasksOnTheCr
it
ical
Pat
h

2.Repl
anSer
ial
Pat
hsToBePar
all
el

3.Ov
erl
apSequent
ial
Tasks

4.Shor
tenTheDur
ati
onOnCr
it
ical
Pat
hAct
ivi
ti
es

5.Shor
tenEar
lyTasks

6.Shor
tenLongestTasks

7.Shor
tenEasi
estTasks

8.Shor
tenTasksThatCostTheLeastToSpeedUp

7.18 Float
sorSlack:Me asur esoff loat
Wehav eseenthatift
hedif
ferencebet weenthemax i
mum ti
meav ail
abl
eandthe
durati
onoftheacti
vit
y,t
hetotalfloat,
is0thentheacti
vi
tyi
scri
tical.Ther
earetwo
otheri
mportantmeasuresoffloat,fr
eefl
oatandindependentf
loat .

Fr
eef
loat

Freefl
oatist
heti
methatanact
ivi
tycoul
dbedel
ayedwi
thoutaf
fect
inganyoft
he
acti
vi
ti
esthatf
oll
ow.

Fr
eef
loat=ESTf
orj
-ESTf
ori
-dur
ati
onofact
ivi
ty

Howev
er,
freef
loatdoesassumet
hatpr
evi
ousact
ivi
ti
esr
unt
oti
me.

I
ndependentf
loat

2
2
Theindependentf
loatgi
vesthet
imethatanact
ivi
tycoul
dbedel
ayedifal
lthepr
ev i
ous
act
ivi
ti
esarecompletedaslat
easpossibl
eandall
thefol
lowi
ngact
ivi
ti
esaretostart
asearl
yaspossibl
e.

I
ndependentf
loat=ESTf
orj
-LSTf
ori
-dur
ati
onofact
ivi
ty

Thedet
ermi
nat
ionoft
otal
,fr
eeandi
ndependentf
loati
sil
l
ust
rat
edi
nFi
gur
e20.
10.

Example:
7.1TasksA,B,C,
…..
H,Iconst
it
uteaproject.Theprecedencerel
ati
onshi
pare
A<D; A<E; B<F; D<F; C<G; C<H; F<I ;
G<I .
Drawanet wor
ktorepresentt
hepr oj
ectandfindthemi ni
mum timeofcompl et
ionof
thepr
ojectwhenti
me, indays,ofeachtaskisasf ol
lows:

Tabl
e7.
2
Task : A B C D E F G H I
Ti
me : 8 10 8 10 16 17 18 14 9
Alsoi
dent
if
ythecri
ti
cal
path.

2
2
Sol
:

Fi
g(7.
1)

Thegivenprocedur eorderrevealsthattherear enopr edecessorstoact

ivit
iesA,BandC
andtheyallstartfrom theini
tialnode.Similarl
y,ther
ear enosuccessort oactiv
iti
esE,H
andIandhencet heyallmer geintot heendnodeofpr oject.Thenetworkobt ai
nedis
showni nfig( 7.1)
Thenodesoft henet workhav ebeennumber edbyusi ngt heFul
kerson’srule.The
acti
vi
tydescri
pt i
onandt imesar ewr itt
enal ongt heactivit
yar r
ows.Todet erminethe
minimum proj
ectcompl et
iontime,letevent1occuratz erotime.Theearl
iestoccurr
ence
ti
me( E)andthel atestoccurr
encet i
me( L)ofeachev entisthencomputed.
E1=0,
E2=E1+t
12=0+8=8,
E3=E1+t
13=0+8=8,
E4=Max.
[0+10,
8+10]=18,
E5=Max.
[18+17,
8+18]=35,
E6=Max.
[8+16,
35+9,
8+14]=44.
Si
mil
arl
y,

L6=E6=44,
L5=L6-t
56=44-9=35,
L4=
L5–t
45=35-17=18,
L3=Mi
n.[44-14,
35-
18]=17,

2
2
 L2=Mi
n.[44-16,
18-
10]=8,
L1=Mi
n.[8-8,
17-8,
18-
10]=0.
TheEandLv
aluesf
oreachev
enthav
ebeenwr
it
tenal
ongt nF
henodesi i
g(7.
2)

FI
G(7.
2)

Thecri
ti
calpat
hisnowdeter
minebyanyoft
hefol
lowi
ngmethod.
Method1.Thenetwor
kanaly
sist
abl
eiscompli
edasbel
ow.

Tabl
e7.
3

Act
ivi
ty Dur
ati
on St
artt
ime Fini
sht
ime Tot
al
Ear
li
est Ear
liest fl
oat
Lat
est Lat
est
1-
2 8 0 0 8 8 0
1-
3 8 0 9 8 17 9
1-
4 10 0 8 10 18 8
2-
4 10 8 8 18 18 0
2-
6 16 8 28 24 44 20
3-
5 18 8 17 26 35 9
3-
6 14 8 30 22 44 22
4-
5 17 18 18 35 35 0
5-
6 9 35 35 44 44 0

2
2
Acti
vi
ti
es1-2,
2-4,
4-5and5-
6hav
ingz
erof
loatar
ethecr
it
ical
act
ivi
ti
esand1-
2-4-
5-6i
s
thecri
ti
cal
path.

Method2.Fori dentif
y i
ngthecr i
ti
calpath,thefol
lowingconditi
onarechecked.I
fan
acti
vit
ysat i
sfi
esallthethreeconditi
ons, i
tiscri
tical
.
(i
).E=Lf orthetail
ev ent.
(i
i) E=Lf ortheheadev ent.
(i
ii
)Ej-Ei=Lj-Li=ti
j.
Acti
v i
ti
es1-2,2-
4,4-
5and5- 6sati
sfythesecondi t
ions.Otheracti
vi
ti
esdonotf ul
fi
lal
l
thethreeconditi
ons.Thecr it
icalpat
his, theref
ore,1-2-
4-5-
6.

Met
hod3.Thev
ari
ouspat
hsandt
hei
rdur
ati
onar
e:

Tabl
e7.
4

Path Durat
ion(day
s)
1-2-6 24
1-2-4-5-6 44
1-4-5-6 36
1-3-5-6 35
1-3-6 22
Pat
h1-
2-4-
5-6,t
helongesti
nti
mei nvolv
ing44day
s,isthecri
ti
cal
pat
h.I
trepr
esent
ed
bybol
dli
nesinfig(7.2).
E=8
L=8

Ex
ampl
e7.
2:Apr
ojectschedul
ehasthef
oll
owi
ngchar
act
eri
sti
cs:
Tabl
e7.
5

Acti
vi
ty Ti
me(
weeks) Acti
vit
y Ti
mes(
weeks)
1-
2 4 5-6 4
1-
3 1 5-7 8
2-
4 1 6-8 1
3-
4 1 7-8 2
3-
5 6 8-10 5

2
2
 4-9 5 9-
10 7
(
i)
.Constr
uctt
henet wor k.
(
ii
).ComputeEandLf oreach,
and
(
ii
i)
.Findt
hecri
ticalpath.

Solut
ion:Thegi
vendatar
esul
tinanetwor
kshowni
nfi
g(7.
5).Thef
igur
esal
ong
thearrowsrepr
esentt
heact
ivi
tyt
imes.

Theear
liestoccurrenceti
me(E)andt
helat
estoccur
renceti
me( L)ofeachev
entar
enow
computedbyempl oyi
ngforwar
dandbackwardpasscalcul
ati
ons.
Inf
orwardpasscomput at
ions,
E1=0,
E2=E1+t
12=0+4=4,
E3=E1+t
13=0+1=1,
E4=Max[ Ei
+t4]
i =Max.
[4+1,
1+1]=5,
i=2,
3
E5=E3+t35=1+6=7,
E6=E5+t
56=7+4=11,
E7=E5+t
57=7+8=15,
E8=Max[
Ei+t
i8]=Max
.[11+1,
15+1]=17,
i=6,
7
E9=E4+t
49=5+5=10,
and
E10=Max[ Ei
+t 10]
i =Max .
[17+5,
10+7]=22,
i
=8,9
Eval
uesar erepr
esentedinfig7. 2.
I
nbackwar dpasscomput at
ions,
L10=E10=22,
L9=L10-t
9,10=22-
7=15,
L8=L10–t
8,10=22-
5=17,
L7=L8–t
78 =17-
2=15,
L6=L8–t
68 =17-
1=16,
L5=Mi
n[Lj
-t ]=Mi
5j n[16-
4,15-
8]=7,
j=6,
7
L4=L9–t
49 =15-
5=10,

2
2
 L3=Mi
n[Lj
–t ]=Mi
3j n[10-4,
7-6]=1,
j=4,
5
L2=L4–t
24 =10-
1=9,
L1=Min[
Lj–t ]=Mi
1j n[9-4, 1-1]=0,j=2,
3
Lval
uesarealsorepresenti
nf i
g(7.
2).
Table7.6
Act
ivi
ty Durati
on St
artti
me Fi
nisht
ime Total
(weeks) Ear li
est Ear
li
est Lat
est fl
oat
Latest
1-
2 4 0 5 4 9 5
1-
3 1 0 0 1 1 0
2-
4 1 4 9 5 10 5
3-
4 1 1 9 2 10 8
3-
5 6 1 1 7 7 0
4-
9 5 5 10 10 15 5
5-
6 4 7 12 11 16 5
5-
7 8 7 7 15 15 0
6-
8 1 11 16 12 17 5
7-
8 2 15 15 17 17 0
8-
10 5 17 17 22 22 0
9-
10 7 10 15 17 22 5

Pat
h1-
3-5-
7-8-
10wi
thpr
ojectdur
ati
on22weeksi
sthecr
it
ical
pat
h.

Example7.
3:Theuti
li
tydataforanetworkaregi
venbel
ow.Determi
nethet
otal
,fr
ee,
i
ndependentandint
erfer
ingfloat
sandidenti
fyt
hecri
ti
calpat
h.
Tabl
e7.
7
Acti
vi
ty 0- 1 1-2 1- 3 2- 4 2- 5 3- 4 3- 6 4- 7 5- 7 6- 7
Durat
ion 2 8 10 6 3 3 7 5 2 8

Sol:Thenetworkdi agram forthegivenpr ojectdataisshowninfig(

1.3).Act
ivi
tydur
ati
onsare
writ
tenalongtheact iv
ityarrows.
Theearli
eststartandl atestfi
nishti
mesoft heacti
vi
tiesarecomputedbyempl oyi
ngthefor
ward
passandbackwar dpasscal culati
onasex plainedi
npr ev
iousexample.Thesetimesare
repr
esentedinnetwor kar oundt herespectivenodes.
Thenetwor kanalysistableisnowconst r
uct ed.

3
3
 Fi
g(7.
3)

Tabl
e7.
8
Act
ivi
ty Dur
ati
on Star
tTi
me Fi
nishti
me Fl
oat
Earl
i
estLatest Ear
li
estLatest Total Fr ee Independent
I
nterf
eri
ng
1 2 3 4 5 6 7 8 9 10
0-
1 2 0 0 2 2 0 0 0 0
1-
2 8 2 8 10 16 6 0 0 6
1-
3 10 2 2 12 12 0 0 0 0
2-
4 6 10 16 16 22 6 0 -
60 6
2-
5 3 10 22 13 25 12 0 -
60 12
3-
4 3 12 19 15 22 7 1 1 6
3-
6 7 12 12 19 19 0 0 0 0
4-
7 5 16 22 21 27 6 6 0 0
5-
7 2 13 25 15 27 12 12 0 0
6-
7 8 19 19 27 27 0 0 0 0

Totalf
loati
stheposi ti
vediff
erencebetweenl at
estandearli
estf
ini
sht
imesorl
atestandear
li
est
st
artti
me.Foract ivi
ty1-2,
Totalf
loat(T.
F)=16-10=8- 2=6.
Si
mi l
arl
y,foracti
v i
ty,say2-5,
Totalfl
oat=25- 13=22-10=12andsoon.
Totalf
loatcal
culati
onsar edepictedi
ncol umn7oft able1.
28.
Freefl
oatacti
vit
y i
-j=T.F.-headeventslack

3
3
 =T.
F.–(
L-E)ofev
entj
.
Thusf
reef
loatofact
ivi
ty0-
1
=0-(L–E)ofev ent1,
=0–(2-2) =0,
Fr eef loatofact ivity1- 2 =6- (16-10)=6- 6=0, etc.
Fr eef loatofv ari
ousact ivit
iesar ecalculatedi ncol umn8oft henetworkanalysistable.
I
ndependentf l
oatofact ivit
yi -j
=F.F–t aileventsl ack=F. F–( L-
E)ofev enti
.
Thusi ndependentf l
oatofact ivi
ty0-1=0–( 0-0)=0,
I
ndependentf l
oatofact ivit
y1- 2=0–(2- 2)=0,
I
ndependentf l
oatofact ivit
y2- 4=0–( 16- 10) =-60andsoon.
I
ndependentf l
oatofv ar
iousact ivit
iesar ecal culatedincolumn9oft hetabl
e.Ifindependent
floatofanact iv
ityisnegat i
v e, i
tistakenaz ero.
I
nt erf eringf l
oatofact i
vi
tyi-j=Max [L.F.timeofi -j–E. S.t
imeofj-k,0]Thus
i
nt erfer ingf l
oatofact i
vi
ty0- 1
=Max.[ L.F.ti
meof0- 1–E. S.t i
meof1- 2or1- 3,0]=Max.[ 2- 2,
0]=0, Int
erfer
ing
floatofact ivi
ty1-2=Max.[ L.F.ti
meof1- 2–E. S.timeof2-4or2- 5,0]
=Max.[ 16- 10,0]=6, etc
Mor econv eni
ently, interferi
ngf loatofanact i
vity=T. F.–F.F.
Thus, int er
feri
ngf loatofact i
v i
ty2-5=12- 0=12, etc.
Al ternat ing, i
nterferingf l
oatofanact i
vity2-5=25- 13=12,etc.
I
nt erf eringf l
oatsofv ari
ousact i
vit
iesar ecal culatedincolumn10oft able7.
8.

Example7.
4:APERTnet workisshowniffi
g1. 4.Theactiv
ityt
imesinday saregi
venalongthe
arr
ows.Thescheduledt
imesf orsomeimportantev ent
saregivenalongthenodes.Deter
mine
thecr
it
icalpat
handprobabil
i
tiesofmeeti
ngt hescheduleddatesforthespeci
fi
edevents.
Tabul
atetheresul
tanddeter
mi nesl
ackforeachev ent.

Fi
g(7.
4)

3
3
Sol
uti
on: Tabl
e7.
9
Acti
vi
ty t
0 t
m t
p t
e 2

1-
2 2 4 12 5.00 2.78
1-
3 2 4 25 7.17 14.69
2-
1 1 1 5 84.83 1.36
3-
4 1 3 6 3.17 0.69
3-
5 1 4 9 4.33 1.78
3-
6 2 5 12 5.67 2.78
3-
7 2 3 5 3.17 0.25
4-
8 5 6 14 7.17 2.25
5-
2 3 6 9 6.00 1.00
6-
9 4 6 10 6.33 1.00
7-10 4 5 7 5.17 0.25
8-11 2 4 6 4.00 0.44
9-12 1 2 4 2.17 0.25
10-12 1 4 8 4.17 1.36
11-12 3 4 11 5.00 1.78

Thear
rowdiagr
am fort
hegi v
endataisshowninfig7.
4.Theex
pectedact
ivi
tyt
imesar
eshown
al
ongthearr
ows.Theearl
iestandl
atestoccur
renceti
mesaswell
astheslackoft
heeventar
e
al
sowrit
tenal
ongthenodes.

Fi
g(7.
5)

3
3
Tabl
e7. 10repr
esentt henetworkanalysi
s.Floatsfortheact
ivi
ti
esi nquest
ionar
e
cal
cul
atedinthelastcolumn.Cr
it
icalpathi
s1-
3-4-8-
11-12andt
heprojectcompl
eti
oni
s
26.
51day s:
Tabl
e7.
10
Act
ivi
ty Dur
ati
on Star
tTi
me Fini
shti
me Total
Earl
i
estLatest Earl
i
estLatest fl
oat

1-2 5.
00 0 11.
68 5.
00 16.
68 11.
68
1-3 7.
17 0 0 7.
17 7.
17 0
2-11 4.
83 5.
00 16.
68 9.
83 21.
51 11.
68
3-4 3.
17 7.
17 7.
17 10.
34 10.
34 0
3-5 4.
33 7.
17 16.
18 11.
50 20.
51 9.
01
3-6 5.
67 7.
17 12.
33 12.
84 18.
00 5.
16
3-7 3.
17 7.
17 14.
00 10.
34 17.
17 6.
83
4-8 7.
17 10.
34 10.
34 17.
51 17.
51 0
5-12 6.
00 11.
50 20.
51 17.
50 26.
51 9.
01
6-9 6.
33 12.
84 18.
01 19.
17 24.
34 5.
17
7-10 5.
17 10.
34 17.
17 15.
51 22.
34 6.
83
8-11 4.
00 17.
51 17.
51 21.
51 21.
51 0
9-12 2.
17 19.
17 24.
34 21.
34 26.
51 5.
17
10-12 4.
17 15.
51 22.
34 19.
68 26.
51 6.
83
11-12 5.
00 21.
51 21.
51 26.
51 26.
51 0

Probabil
it
yofcompl
eti
ngt
hepr
ojecti
ntheschedul
edcompl
eti
ont
imeof24day
s(si
nce
Ts(12)=24):

Z= 24-
26.
51 =-
2.55_=-
0.5634.

 p(Ts≤24)=1-valueofprobabi
li
tyforZ=0.5634=1-0.
7146=29.
54%.
Pr
obabi
l
ityt
hatev
ent3wil
l occuronthescheduleddat
e:
Ts(
3)=4,
E=L=7. 17

 Z=4-
7.17=-
0.8271

p=1-valueofpr obabil
ityf
orZ=0. 8271=1-0. 7956=20.44%
Probabi
l
ityofmeet i
ngschedul eddateforevent5:
Theearl
iestoccur
rencet imeofev ent5is11.50,whil
etheschedul
edt
imeis12.
Event
5isnotoncr i
ti
calpat handhencei t
soccurrencecanbedelayedby9day
s.
Vari
anceofpath1- 3-5=14.69+1.78=16. 47.

3
3
 Ts=12, Ts=12
E=11.
50 L=20.
51.
 Z =12–11.
50=0.
123  Z=12–20.51=-
2.1


Pr
obabi
l
ity=54.
89% 
Pr
obabi
l
ity=1–0.
982=0.
018=1.
8%

Example7.5:I
nt hePERTnet workshowninf i
g7. 6,theacti
vit
yti
meest i
mat es(
in
weeks)aregivenal ongthear r
ows.Iftheschedul
edcompleti
onti
mei s23weeks,
cal
culat
et hel
at estpossibleoccur
renceti
mesoft heevent
s.Cal
culat
etheslackfor
eacheventandi dentif
ythecr i
ti
calpat
h.Whati st
heprobabi
li
tyt
hatproj
ectwillbe
completedont heschedul eddate?.

Fi
g(7.6)
Sol
:Theex
pect
edt
imeoft
heact
ivi
ti
esandtheirv
ari
ancesar
ecomput
edbel
ow.

Act
ivi
ty t
0 t
m t
p t
e 2

1-
2 3 3 3 3 0
2-
3 3 6 9 6 1
2-
4 2 4 6 4 4/
9
3-
5 4 6 8 6 4/
9
4-
6 4 6 8 6 4/
9
5-
6 0 0 0 0 0
5-
7 3 4 5 4 1/
9
6-
7 2 5 8 5 1
Table7.
11
Theearli
estoccur
rencetimesoftheev ent
shav ebeencomput edont henetworkoffig
7.6,takingtheearl
iestt
imeofev ent1asz ero.Theearli
estoccurr
encetimeofev ent7
i
s20.Butt hescheduledcomplet
iontimeoft heproj
ectis23weeksandhencet helatest
occurr
encet i
mesoft heeventshav ebeencomput edtakingL(7)=23.Slacksfort he
eventhavebeenshownal ongthenodes.Pat h1-2-
3-5-
6-7isthecri
ti
calpath.

3
3
 Fi
g(7.
7)

Probabi
l
ityofcomplet
ingt
hepr
ojectonschedul
edate:
T=23weeks, E=20weeks.
Vari
enceofthecri
ti
calpat
h=0=1+4/9=0+1=22/9=2.
444


Z=1.
92.


Probabi
li
ty=97.
26%
Example7.6:Thefoll
owingt
abl
egivesdat
aonnor
mal t
imeandcostandcr
ashti
me
andcostforaproject
.
Acti
vi
ty Normal Crash
Ti
me( days) Time(
Days) Cost
Cost(
Rs.) (Rs.
)
1-
2 6 60 4 100
1-
3 4 60 2 200
2-
4 5 50 3 150
2-
5 3 45 1 65
3-
4 6 90 4 200
4-
6 8 80 4 300
5-
6 4 40 2 100
6-
7 3 45 2 80
470

3
3
TheindirectcostperdayisRs.10.
(i
).Drawt henetworkforproj
ect.
(i
i
).Findt hecri
ti
calpath.
(i
i
i).Determineminimum totalt
imeandcor
respondi
ngcost
.

Sol
uti
on:Fir
stt
hecostsl
opeforeachacti
vi
tyandnor
mal
dir
ectcostoft
hepr
ojecti
s
cal
cul
ated.Thi
sisshowni
nthetablebel
ow.

Acti
vit
y 1-
2 1-
3 2-
4 2-
5 3-
4 4-
6 5-
6 6-
7
Costslope 20 70 50 10 55 55 30 35
(Rs.
/day)

(
i)
.Next
,thenet
wor
kisdr
awandcr
it
ical
pat
hisdet
ermi
ned.Thi
sisshowni
nfi
g1.
8.

Fi
g(7.
8)

(
ii
).Thecr i
ti
cal pathis1-2-
4-6-7.
(
ii
i)
.Nor mal durati
on=22day s.
Normal cost=Rs.(470+22*10)
=Rs.690.
Nowr epresentt henetworkont i
me-scal
eddi
agr
am.Thi
sisshowni
nfi
g(7.
8).
E=22
L=22

3
3