Post Graduation Adventure

7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 /4 4 P ag e 1 28 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d .

C H APT ER F IV E E stim atin g P ro je ct T im es a n d C osts E stim atin g P ro je ct T im es a n d C osts F acto rs I n flu en cin g t h e Q uality o f E stim ate s E stim atin g G uid elin es f o r T im es, C osts , a n d R eso urc es T op-D ow n v ers u s B otto m -U p E stim atin g M eth ods f o r E stim atin g P ro je ct T im es a n d C osts L ev el o f D eta il T ypes o f C osts R efin in g E stim ate s C re atin g a D ata b ase f o r E stim atin g S um mary A ppen dix 5 .1 : L earn in g C urv es f o r E stim atin g 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 /4 4 P ag e 1 29 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . P ro je ct e stim atio n i s i n deed a y a rd sti c k f o r p ro je ct c o st c o ntr o l. A nd i f t h e y a rd stic k i s f a ulty , y o u s ta rt o n t h e “ w ro ng f o ot.” … w e e xh ort y o u n ot t o u ndere stim ate t h e e stim ate . * G iv en th e u rg en cy to s ta rt w ork o n th e p ro je ct, m an ag ers s o m etim es m in im iz e o r a v oid th e e ff o rt to fo llo w t h ro ugh o n e stim atin g p ro je ct tim e a n d c o st. T his a ttitu de is a h uge m is ta k e a n d c o stly . T here a re im porta n t r e aso ns to m ak e th e e ff o rt a n d in cu r th e c o st o f e stim atin g f o r y our p ro je ct. E xhib it 5 .1 s u m mariz es s o m e k ey r e aso ns. E stim atin g is th e pro cess of fo re castin g or ap pro xim atin g th e tim e an d co st of co m ple tin g pro je ct d eliv era b le s. E stim atin g p ro cesse s a re f re q uen tly c la ssif ie d a s t o p-d ow n a n d b ott o m -u p. T op-d ow n e stim ate s a re u su ally d one b y s e n io r m an ag em en t. M an ag em en t w ill o fte n d eriv e e stim ate s f ro m a n alo gy, g ro up c o nse n su s, o r m ath em atic al r e la tio n sh ip s. B otto m -u p e stim ate s a re t y pic ally p erfo rm ed b y t h e p eo ple w ho a re d oin g t h e w ork .

T heir e stim ate s a re b ase d o n e stim ate s o f e le m en ts f o und i n t h e w ork b re ak dow n s tr u ctu re . A ll p ro je c t sta k eh o ld ers p re fe r accu ra te co st an d tim e estim ate s, b ut th ey als o u nders ta n d th e in here n t u ncerta in ty in all pro je cts . In accu ra te estim ate s le ad to fa ls e ex pecta tio ns an d co nsu m er dis sa tis fa ctio n.

A ccu ra cy is im pro ved w ith g re ate r e ff o rt, b ut is it w orth th e tim e a n d c o st— estim atin g c o sts m oney ! P ro je ct e stim atin g b eco m es a t r a d e-o ff , b ala n cin g t h e b en efits o f b ette r a ccu ra cy a g ain st t h e c o sts f o r s e cu rin g i n cre ase d a ccu ra cy. C ost, tim e, a n d b udget e stim ate s a re th e lif e lin e fo r c o ntr o l; th ey s e rv e a s th e s ta n dard fo r c o m paris o n o f a ctu al a n d p la n t h ro ughout t h e l if e o f t h e p ro je ct. P ro je ct s ta tu s r e p orts d ep en d o n r e lia b le e stim ate s a s t h e m ajo r i n put f o r m easu rin g v aria n ces a n d ta k in g c o rre ctiv e a ctio n. I d eally , th e p ro je ct m an ag er, a n d in m ost c ase s th e c u sto m er, w ould p re fe r t o h av e a d ata b ase o f d eta ile d s c h ed ule a n d c o st e stim ate s f o r e v ery w ork p ack ag e i n t h e p ro je ct. R eg re tta b ly , s u ch d eta ile d d ata g ath erin g i s n ot a lw ay s p ossib le o r p ra ctic al a n d o th er m eth ods a re u se d t o d ev elo p p ro je ct e stim ate s.

E X H IB IT 5 .1 W hy E stim atin g T im e a n d C ost A re I m porta n t E stim ate s a re n eed ed t o s u pport g ood d ecis io ns.

E stim ate s a re n eed ed t o s c h ed ule w ork .

E stim ate s a re n eed ed t o d ete rm in e h ow l o ng t h e p ro je ct s h ould t a k e a n d i ts c o st.

E stim ate s a re n eed ed t o d ete rm in e w heth er t h e p ro je ct i s w orth d oin g.

E stim ate s a re n eed ed t o d ev elo p c ash f lo w n eed s.

E stim ate s a re n eed ed t o d ete rm in e h ow w ell t h e p ro je ct i s p ro gre ssin g.

E stim ate s a re n eed ed t o d ev elo p t im e-p hase d b udgets a n d e sta b lis h t h e p ro je ct b ase lin e. 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 /4 4 P ag e 1 30 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . F acto rs I n flu en cin g t h e Q uality o f E stim ate s A ty pic al sta te m en t in th e fie ld is th e d esir e to “h av e a 9 5 p erc en t p ro bab ility o f m eetin g tim e an d co st e stim ate s.” P ast e xp erie n ce is a g ood s ta rtin g p oin t f o r d ev elo pin g tim e a n d c o st e stim ate s. B ut p ast e x perie n ce e stim ate s m ust a lm ost a lw ay s b e re fin ed b y o th er c o nsid era tio ns to re ach th e 9 5 p erc en t p ro bab ility le v el.

F acto rs re la te d to th e u niq uen ess o f th e p ro je ct w ill h av e a str o ng in flu en ce o n th e a ccu ra cy o f e stim ate s.

P ro je ct, p eo ple , a n d e x te rn al f a cto rs a ll n eed to b e c o nsid ere d to im pro ve q uality o f e stim ate s f o r p ro je ct tim es a n d c o sts .

P la n nin g H oriz o n T he q uality o f th e e stim ate d ep en ds o n th e p la nnin g h oriz o n; e stim ate s o f c u rre n t e v en ts a re c lo se to 1 00 p erc en t accu ra te b ut are re d uced fo r m ore d is ta n t ev en ts . F or ex am ple , co st estim ate s fo r a p arty y ou are o rg an iz in g th is w eek en d s h ould b e m uch m ore a ccu ra te th an th e e stim ate s f o r a w ed din g th at w ill ta k e p la ce in t h re e m on th s. T he a ccu ra cy o f tim e a n d c o st e stim ate s s h ould im pro ve a s y ou m ove f ro m th e c o ncep tu al p hase t o th e p oin t w here in div id ual w ork p ack ag es a re d efin ed . L ong-d ura tio n p ro je cts in cre ase th e u ncerta in ty in e stim ate s.

P ro je ct D ura tio n T im e to im ple m en t n ew t e ch nolo gy h as a h ab it o f e x pan din g in a n in cre asin g, n onlin ear fa sh io n. S om etim es p oorly w ritte n s c o pe s p ecif ic atio ns f o r n ew t e ch nolo gy r e su lt i n e rro rs i n e stim atin g t im es a n d c o sts .

P eo ple T he p eo ple fa cto r c an in flu en ce th e q uality o f tim e a n d c o st e stim ate s. F or e x am ple , a c cu ra cy o f e stim ate s d ep en ds o n th e sk ills o f th e p eo ple m ak in g th e estim ate s. H ow fa m ilia r are th ey w ith th e ta sk th ey are e stim atin g?

P ro je ct S tr u ctu re a n d O rg an iz a tio n W hic h p ro je ct s tr u ctu re is c h ose n to m an ag e th e p ro je ct w ill in flu en ce tim e a n d c o st e stim ate s. O ne o f th e m ajo r a d van ta g es o f a d ed ic ate d p ro je ct te am d is c u sse d e arlie r is th e s p eed g ain ed f ro m c o ncen tr a te d f o cu s a n d l o caliz ed p ro je ct d ecis io ns. T his s p eed c o m es a t a n a d ditio nal c o st o f ty in g u p p ers o nnel f u ll tim e. C onvers e ly , p ro je cts o p era tin g in a m atr ix en vir o nm en t m ay re d uce co sts b y m ore eff ic ie n tly sh arin g p ers o nnel acro ss p ro je cts b ut m ay t a k e l o nger t o c o m ple te s in ce a tte n tio n i s d iv id ed a n d c o ord in atio n d em an ds a re h ig her.

P addin g E stim ate s I n s o m e c ase s p eo ple a re i n clin ed t o p ad e stim ate s. F or e x am ple , i f y ou a re a sk ed h ow l o ng i t t a k es y ou t o d riv e t o th e a ir p ort, y ou m ig ht g iv e a n a v era g e tim e o f 3 0 m in ute s, a ssu m in g a 5 0/5 0 c h an ce o f g ettin g th ere in 3 0 m in ute s. If y ou a re a sk ed th e fa ste st y ou c o uld p ossib ly g et th ere , y ou m ig ht re d uce th e d riv in g tim e to 2 0 m in ute s. F in ally , if y ou a re a sk ed h ow lo ng th e d riv e w ould ta k e if y ou a b so lu te ly h ad to b e th ere to m eet w ith t h e p re sid en t, it is lik ely y ou w ould in cre ase th e e stim ate to s a y 5 0 m in ute s to e n su re n ot b ein g la te . In w ork s itu atio ns w here y ou a re a sk ed fo r tim e a n d c o st e stim ate s, m ost o f u s a re in clin ed to a d d a little p ad din g to i n cre ase th e p ro bab ility a n d re d uce th e ris k o f b ein g la te . If e v ery one a t a ll le v els o f th e p ro je ct a d ds a little p ad din g t o r e d uce r is k , t h e p ro je ct d ura tio n a n d c o st 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 4 /4 4 P ag e 1 31 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . a re s e rio usly o vers ta te d . T his p hen om en on c au se s s o m e m an ag ers o r o w ners to c all f o r a 1 0–15 p erc en t c u t in tim e a n d/o r c o st fo r th e p ro je ct. O f c o urs e th e n ex t tim e th e g am e is p la y ed , th e p ers o n e stim atin g c o st a n d/o r tim e w ill p ad th e e stim ate to 2 0 p erc en t o r m ore . C le arly su ch g am es d efe at c h an ces f o r r e alis tic e stim ate s, w hic h i s w hat i s n eed ed t o b e c o m petitiv e.

O rg an iz a tio n C ultu re O rg aniz a tio n c u ltu re c an s ig nif ic an tly in flu en ce p ro je ct e stim ate s. In s o m e o rg an iz atio ns p ad din g e stim ate s is t o le ra te d an d ev en priv ate ly en co ura g ed . O th er org an iz atio ns pla ce a pre m iu m on accu ra cy an d str o ngly d is c o ura g e estim atin g gam esm an sh ip . O rg an iz atio ns vary in th e im porta n ce th ey atta ch to estim ate s. T he p re v ailin g b elie f in so m e o rg an iz atio ns is th at d eta ile d e stim atin g ta k es to o m uch tim e a n d is n ot w orth th e e ff o rt o r th at it's im possib le to p re d ic t th e fu tu re . O th er o rg an iz atio ns su bsc rib e to th e b elie f th at a ccu ra te e stim ate s a re th e b ed ro ck o f e ff e ctiv e p ro je ct m an ag em en t. O rg an iz atio n c u ltu re sh ap es e v ery d im en sio n o f p ro je ct m an ag em en t; e stim atin g i s n ot i m mune t o t h is i n flu en ce.

O th er F acto rs F in ally , n onpro je ct fa cto rs c an im pact tim e a n d c o st e stim ate s. F or e x am ple , e q uip m en t d ow n-tim e c an a lte r t im e e stim ate s. N atio nal h olid ay s, v aca tio ns, a n d le g al lim its c an in flu en ce p ro je ct e stim ate s. P ro je ct p rio rity c an i n flu en ce r e so urc e a ssig nm en t a n d i m pact t im e a n d c o st. P ro je ct e stim atin g is a c o m ple x p ro cess. T he q uality o f tim e a n d c o st e stim ate s c an b e im pro ved w hen th ese v aria b le s a re c o nsid ere d in m ak in g th e e stim ate s. E stim ate s o f tim e a n d c o st to geth er a llo w th e m an ag er to d ev elo p a tim e-p hase d b udget, w hic h is im pera tiv e fo r p ro je ct c o ntr o l. B efo re d is c u ssin g m acro a n d m ic ro e stim atin g m eth ods fo r tim es an d co sts , a re v ie w o f estim atin g g uid elin es w ill re m in d u s o f so m e o f th e i m porta n t “ ru le s o f t h e g am e” t h at c an i m pro ve e stim atin g.

E stim atin g G uid elin es f o r T im es, C osts , a n d R eso urc es M an ag ers re co gniz e tim e, c o st, a n d re so urc e e stim ate s m ust b e a ccu ra te if p ro je ct p la n nin g, s c h ed ulin g, a n d c o ntr o llin g a re to b e e ff e ctiv e. H ow ev er, th ere is s u bsta n tia l e v id en ce s u ggestin g p oor e stim ate s a re a m ajo r c o ntr ib uto r t o p ro je cts t h at h av e f a ile d . T here fo re , e v ery e ff o rt s h ould b e m ad e t o s e e t h at i n itia l e stim ate s a re a s a ccu ra te a s p ossib le s in ce th e c h oic e o f n o e stim ate s le av es a g re at d eal to lu ck a n d is n ot p ala ta b le to s e rio us p ro je ct m an ag ers . E ven th ough a p ro je c t h as n ev er b een d one b efo re , a m an ag er c an f o llo w s e v en g uid elin es to d ev elo p u se fu l w ork p ack ag e e stim ate s. 1 . R esp on sib ilit y . A t th e w ork p ack ag e le v el, e stim ate s s h ould b e m ad e b y th e p ers o n(s ) m ost f a m ilia r w ith t h e ta sk . D ra w o n th eir e x pertis e ! E xcep t f o r s u perte ch nic al ta sk s, th ose r e sp onsib le f o r g ettin g th e jo b d one o n s c h ed ule a n d w ith in b udget a re u su ally fir s t- lin e s u perv is o rs o r te ch nic ia n s w ho a re e x perie n ced a n d fa m ilia r w ith th e ty pe o f w ork in volv ed . T hese p eo ple w ill n ot h av e so m e p re co nceiv ed , im pose d d ura tio n fo r a d eliv era b le in m in d. T hey w ill g iv e a n e stim ate b ase d o n e x perie n ce a n d b est ju dgm en t. A s e co ndary b en efit o f u sin g th ose re sp onsib le is th e h ope th ey w ill “b uy in ” to se ein g th at th e estim ate m ate ria liz es w hen th ey i m ple m en t t h e w ork p ack ag e. I f t h ose i n volv ed a re n ot c o nsu lte d , i t w ill b e d if f ic u lt t o h old t h em r e sp onsib le f o r f a ilu re t o a ch ie v e t h e e stim ate d t im e. 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 5 /4 4 P ag e 1 32 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . F in ally , d ra w in g o n th e ex pertis e o f te am m em bers w ho w ill b e re sp onsib le h elp s to b uild c o m munic atio n c h an nels e arly . 2 . U se s e v era l p eo p le t o e stim ate . I t i s w ell k now n t h at a c o st o r t im e e stim ate u su ally h as a b ette r c h an ce o f b ein g re aso nab le a n d re alis tic w hen s e v era l p eo ple w ith re le v an t e x perie n ce a n d/o r k now le d ge o f th e ta sk a re u se d (s o m etim es c alle d “ cro w dso urc in g”). T ru e, p eo ple b rin g d if f e re n t b ia se s b ase d o n th eir e x perie n ce. B ut d is c u ssio n o f th e in div id ual d if f e re n ces in th eir e stim ate le ad s to c o nse n su s a n d te n ds to e lim in ate e x tr e m e e stim ate e rro rs . 3 . N orm al c o n dit io n s. W hen ta sk tim e, co st, an d re so urc e estim ate s are d ete rm in ed , th ey are b ase d o n c erta in a ssu m ptio ns. E stim ate s s h ould b e b ase d o n n orm al c o nditio ns, e ffic ie n t m eth ods, a nd a n orm al le vel o f r e so urc es . N orm al c o nditio ns a re s o m etim es d if f ic u lt to d is c ern , b ut it is n ecessa ry to h av e a c o nse n su s in th e o rg an iz atio n a s to w hat n orm al c o nditio ns m ean in th is p ro je ct. I f th e n orm al w ork day is e ig ht h ours , th e tim e e stim ate s h ould b e b ase d o n a n e ig ht- h our d ay. S im ila rly , i f t h e n orm al w ork day i s t w o s h if ts , t h e t im e e stim ate s h ould b e b ase d o n a tw o-s h if t w ork day. A ny tim e e stim ate s h ould re fle ct e ff ic ie n t m eth ods fo r th e re so urc es n orm ally a v aila b le . T he t im e e stim ate s h ould r e p re se n t t h e n orm al l e v el o f r e so urc es— peo ple o r e q uip m en t. F or e x am ple , if th re e p ro gra m mers a re a v aila b le f o r c o din g o r tw o r o ad g ra d ers a re a v aila b le f o r r o ad c o nstr u ctio n, t im e a n d c o st e stim ate s s h ould b e b ase d o n th ese n orm al le v els o f r e so urc es u nle ss it is a n tic ip ate d th e p ro je ct w ill c h an ge w hat is c u rre n tly v ie w ed a s “ n orm al.” In a d ditio n, p ossib le c o nflic ts in d em an d fo r re so urc es o n p ara lle l o r c o ncu rre n t a ctiv itie s s h ould n ot b e c o nsid ere d a t th is s ta g e. T he n eed fo r a d din g re so urc es w ill b e e x am in ed w hen r e so urc e s c h ed ulin g i s d is c u sse d i n a l a te r c h ap te r. 4 . T im e u nit s . S pecif ic tim e u nits to u se s h ould b e s e le cte d e arly in th e d ev elo pm en t p hase o f th e p ro je ct n etw ork . A ll ta sk tim e e stim ate s n eed c o nsis te n t tim e u nits . E stim ate s o f tim e m ust c o nsid er w heth er n orm al t im e is re p re se n te d b y c ale n dar d ay s, w ork day s, w ork w eek s, p ers o n d ay s, s in gle s h if t, h ours , m in ute s, e tc . In p ra ctic e t h e u se o f w ork day s i s t h e d om in an t c h oic e f o r e x pre ssin g t a sk d ura tio n . H ow ev er, i n p ro je cts s u ch a s a h eart tr a n sp la n t o pera tio n, m in ute s p ro bab ly w ould b e m ore a p pro pria te a s a tim e u nit. O ne s u ch p ro je ct th at u se d m in ute s a s th e tim e u nit w as th e m ovem en t o f p atie n ts f ro m a n o ld h osp ita l to a n e le g an t n ew o ne a cro ss t o w n. S in ce th ere w ere s e v era l lif e -e n dan gerin g m oves, m in ute s w ere u se d to e n su re p atie n t s a fe ty s o p ro per e m erg en cy l if e -s u pport s y ste m s w ould b e a v aila b le i f n eed ed . T he p oin t i s , n etw ork a n aly sis r e q uir e s a s ta n dard u nit o f tim e. W hen c o m pute r p ro gra m s a llo w m ore th an o ne o ptio n, so m e n ota tio n sh ould b e m ad e o f a n y v aria n ce f ro m th e s ta n dard u nit o f tim e. I f th e s ta n dard u nit o f tim e is a f iv e-d ay w ork w eek a n d th e e stim ate d a ctiv ity d ura tio n i s i n c ale n dar d ay s, i t m ust b e c o nverte d t o t h e n orm al w ork w eek . 5 . I n dep en den ce. E stim ato rs s h ould tr e at e ach ta sk a s in dep en den t o f o th er ta sk s th at m ig ht b e in te g ra te d b y t h e W BS. U se o f fir s t- lin e m an ag ers u su ally re su lts in co nsid erin g ta sk s in dep en den tly ; th is is g ood. T op m an ag ers a re p ro ne to a g gre g ate m an y ta sk s in to o ne tim e e stim ate a n d th en d ed uctiv ely m ak e th e in div id ual t a sk t im e e stim ate s a d d t o t h e t o ta l. I f t a sk s a re i n a c h ain a n d p erfo rm ed b y t h e s a m e g ro up o r d ep artm en t, i t i s b est n ot to a sk fo r a ll th e tim e e stim ate s in th e se q uen ce a t o nce to a v oid th e te n den cy fo r a p la n ner o r a s u perv is o r to lo ok a t th e w hole p ath a n d tr y to a d ju st in div id ual ta sk tim es in th e s e q uen ce to m eet a n a rb itr a ry i m pose d s c h ed ule o r s o m e r o ugh “ g uesstim ate ” o f th e to ta l tim e f o r th e w hole p ath o r s e g m en t o f th e p ro je ct.

T his te n den cy d oes n ot r e fle ct th e u ncerta in tie s o f in div id ual a ctiv itie s a n d g en era lly r e su lts in o ptim is tic ta sk t im e e stim ate s. I n s u m mary , e ach t a sk t im e e stim ate s h ould b e c o nsid ere d i n dep en den tly o f o th er a ctiv itie s. 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 6 /4 4 P ag e 1 33 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . S N APSH OT F R O M P R A CTIC E R ed ucin g E stim atin g E rro rs * C om ple x ity is th e m ajo r so urc e o f estim ati n g erro r, sa y s K erry W ills , P ro je ct M an ag em en t S r. D ir e cto r at th e h ealth care s e rv ic es o rg an iz atio n C ig na, in H artf o rd , C onnectic u t. “ P ro je ct m an ag ers c an not p ossib ly b e e x perts in a ll a re as a n d th ere fo re n eed to r e ly o n th e s ta k eh old ers f o r th eir e x pertis e w hen e stim atin g,” W illis n ote s. T o m in im iz e e rro rs h e r e co m men ds t r e atin g e stim atin g a s a l iv in g p ro cess a n d n ot a o ne-tim e e v en t. H e f o llo w s t h e s a m e a p pro ach o n a ll o f h is p ro je cts :

1 . I d en tif y a ll o f t h e s ta k eh old ers b ase d o n t h e s c o pe o f t h e p ro je ct a n d o rg an iz atio nal h is to ry . 2 . I n volv e t h e s ta k eh old ers w hen c re a tin g t h e e stim ate s. “ Y ou c an 't h old p eo ple a cco unta b le f o r e stim ate s t h ey d id n't h elp c re ate ,” W illis s a y s. 3 . A ggre g ate t h e e stim ate s b y c o m parin g s e v era l m od els ( re so urc e b ase d , p ara m etr ic , e tc .) . 4 . M an age t h e p ro je ct a gain st t h e e stim ate s. T his i n clu des m ak in g a d ju stm en ts b ase d o n c h an ges i n p ro je ct s c o pe. 5 . T ra ck p ro je cts c lo se ly u sin g t o ols s u ch a s e arn ed v alu e t o g au ge p ro gre ss t o w ard e stim ate s. 6 . T ra ck a ctu al c o sts a n d t im e a t a g ra n ula r l e v el t o r e calib ra te t h e m odel f o r f u tu re p ro je cts . “ T he i n itia l e stim ate c o uld b e p erfe ct, b ut i f i t i s n o t m an ag ed , t h en t h e e n d r e su lt w ill b e b ad a n d p eo ple w ill p oin t t o t h e e stim atin g p ro cess,” W ills a rg ues.

* S . S w an so n, “ E stim atin g E rro rs ,” P M Netw ork , O cto ber 2 011 , p p. 6 2–66. 6 . C on tin gen cie s. W ork p acka ge e stim ate s sh ould n ot in clu de a llo w ances fo r c o ntin gen cie s. T he e stim ate s h ould a ssu m e n orm al o r a v era g e c o nditio ns e v en th ough e v ery w ork p ack ag e w ill n ot m ate ria liz e a s p la n ned .

F or th is re aso n to p m an ag em en t n eed s to c re ate a n e x tr a fu nd fo r c o ntin gen cie s th at c an b e u se d to c o ver u nfo re se en e v en ts . 7 . A ddin g r is k a sse ssm en t to th e e stim ate h elp s to a void s u rp ris e s to s ta k eh old ers. It is o bvio us s o m e t a sk s c arry m ore tim e a n d c o st r is k s th an o th ers . F or e x am ple , a n ew te ch nolo gy u su ally c arrie s m ore tim e a n d c o st ris k s th an a p ro ven p ro cess. S im ply id en tif y in g th e d eg re e o f ris k le ts s ta k eh old ers c o nsid er a lte rn ativ e m eth ods a n d a lte r p ro cess d ecis io ns. A s im ple b re ak dow n b y o ptim is tic , m ost lik ely , a n d p essim is tic fo r ta sk t im e c o uld p ro vid e v alu ab le i n fo rm atio n r e g ard in g t im e a n d c o st. S ee C hap te r 7 f o r f u rth er d is c u ssio n o f p ro je ct r is k .

W here a p plic ab le , t h ese g uid elin es w ill g re atly h elp t o a v oid m an y o f t h e p itf a ll s f o und s o o fte n i n p ra ctic e. S ee S nap sh ot f ro m P ra ctic e: R ed ucin g E stim atin g E rro rs f o r a s im ila r s e t o f g uid elin es.

T op-D ow n v ers u s B otto m -U p E stim atin g S in ce e stim atin g e ff o rts c o st m oney, th e tim e a n d d eta il d ev ote d to e stim atin g is a n im porta n t d ecis io n. Y et, w hen e stim atin g i s c o nsid ere d , y ou a s a p ro je ct m an ag er m ay h ear s ta te m en ts s u ch a s t h ese : 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 7 /4 4 R ough o rd er o f m agnitu de i s g ood e n ough. S pen din g t im e o n d eta ile d e stim atin g w aste s m oney.

T im e i s e very th in g; o ur s u rv iv a l d ep en ds o n g ettin g t h ere f ir st! T im e a nd c o st a ccu ra cy i s n ot a n i s su e.

T he p ro je ct i s i n te rn al. W e d on't n eed t o w orry a bout c o st.

T he p ro je ct i s s o s m all, w e d on't n eed t o b oth er w ith e stim ate s. J u st d o i t. 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 8 /4 4 P ag e 1 34 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . T A BLE 5 .1 C on dit io n s f o r P re fe rrin g T op -D ow n o r B otto m -U p T im e a n d C ost E stim ate s H ow ev er, th ere a re s o und r e aso ns f o r u sin g to p-d ow n o r b otto m -u p e stim ate s. T ab le 5 .1 d ep ic ts c o nditio ns th at s u ggest w hen o ne a p pro ach i s p re fe rre d o ver a n oth er. T op -d ow n e stim ate s u su ally a re d eriv ed f ro m s o m eo ne w ho u se s e x perie n ce a n d/o r i n fo rm atio n t o d ete rm in e t h e p ro je ct d ura tio n a n d to ta l c o st. H ow ev er, th ese e stim ate s a re s o m etim es m ad e b y to p m an ag ers w ho h av e v ery l ittle k now le d g e o f t h e c o m ponen t a ctiv itie s u se d t o c o m ple te t h e p ro je ct. F or e x am ple , a m ay or o f a m ajo r c ity m ak in g a s p eech n ote d t h at a n ew l a w b uild in g w ould b e c o nstr u cte d a t a c o st o f $ 23 m illio n a n d w ould b e r e ad y fo r o ccu pan cy in tw o an d o ne-h alf y ears . A lth ough th e m ay or p ro bab ly ask ed fo r an estim ate fro m s o m eo ne, t h e e stim ate c o uld h av e c o m e f ro m a l u nch eo n m eetin g w ith a l o cal c o ntr a cto r w ho w ro te a n e stim ate ( g uesstim ate ) o n a n ap kin . T his i s a n e x tr e m e e x am ple , b ut i n a r e la tiv e s e n se t h is s c en ario i s f re q uen tly p la y ed o ut in p ra ctic e. S ee S nap sh ot f ro m P ra ctic e: C ouncil F um es, f o r a n oth er e x am ple o f th is . T he q uestio n a ctu ally i s , d o th ese e stim ate s r e p re se n t lo w -c o st, e ffic ie n t m eth ods ? S eld om . T he fa ct th at th e e stim ate c am e fro m th e t o p c an i n flu en ce p eo ple r e sp onsib le t o “ d o w hat i t t a k es t o m ak e t h e e stim ate .” I f p ossib le a n d p ra ctic al, y ou w an t t o p u sh t h e e stim atin g p ro cess d ow n t o t h e w ork p ack ag e l e v el f o r b otto m - u p e stim ate s th at e sta b lis h lo w -c o st, e ff ic ie n t m eth ods. T his p ro cess c an ta k e p la ce a fte r th e p ro je ct h as b een d efin ed in d eta il. G ood se n se su ggests p ro je ct e stim ate s sh ould c o m e fro m th e p eo ple m ost k now le d geab le a b out th e e stim ate n eed ed . T he u se o f s e v era l p eo ple w ith re le v an t e x perie n ce w ith th e ta sk c an im pro ve th e t im e a n d c o st e stim ate . T he b otto m -u p a p pro ach a t th e w ork p ack ag e le v el c an se rv e a s a c h eck o n c o st e le m en ts in th e W BS b y ro llin g u p th e w ork p ack ag es a n d a sso cia te d c o st a cco unts to m ajo r d eliv era b le s.

S im ila rly , re so urc e re q uir e m en ts c an b e c h eck ed . L ate r, th e tim e, re so urc e, a n d c o st e stim ate s fro m th e w ork p ack ag es c an b e c o nso lid ate d in to tim e-p hase d n etw ork s, re so urc e sc h ed ule s, a n d b udgets th at a re u se d fo r c o ntr o l. T he b otto m -u p a p pro ach a ls o p ro vid es th e c u sto m er w ith a n o pportu nity to c o m pare th e lo w -c o st, e ff ic ie n t m eth od a p pro ach w ith a n y im pose d r e str ic tio ns. F or e x am ple , if th e p ro je ct c o m ple tio n d ura tio n is im pose d a t t w o y ears a n d y our b otto m -u p a n aly sis t e lls y ou t h e p ro je ct w ill t a k e t w o a n d o n e-h alf y ears , t h e c lie n t c an n ow c o nsid er th e tr a d e-o ff o f th e lo w -c o st m eth od v ers u s c o m pre ssin g th e p ro je ct to tw o y ears — or in ra re c ase s c an celin g th e p ro je ct. S im ila r tr a d e-o ff s can b e co m pare d fo r d if f e re n t le v els o f re so urc es o r in cre ase s in t e ch nic al p erfo rm an ce. T he a ssu m ptio n i s a n y m ovem en t a w ay f ro m t h e l o w -c o st, e ff ic ie n t m eth od w ill i n cre ase c o sts — e.g ., o vertim e. T he p re fe rre d a p pro ach in d efin in g th e p ro je ct is to m ak e ro ugh to p-d ow n e stim ate s, d ev elo p th e W BS/O BS, m ak e b otto m -u p e stim ate s, d ev elo p s c h ed ule s a n d b u dgets , a n d re co ncile d if f e re n ces b etw een to p-d ow n a n d b otto m -u p e stim ate s. H opefu lly , th ese s te p s w ill b e d one b efo re fin al n eg otia tio n w ith e ith er a n in te rn al o r e x te rn al c u sto m er. In c o nclu sio n, th e id eal a p pro ach is fo r th e p ro je ct m an ag er to a llo w e n ough tim e fo r b oth th e to p-d ow n a n d b otto m -u p e stim ate s to b e w ork ed o ut s o a c o m ple te p la n b ase d o n r e lia b le estim ate s can be off e re d to th e cu sto m er. In th is w ay fa ls e ex pecta tio ns are m in im iz ed fo r all s ta k eh old ers a n d n eg otia tio n i s r e d uced . 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 9 /4 4 P ag e 1 35 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . S N APSH OT F R O M P R A CTIC E C ouncil F um es a s T ra m T ale U nfo ld s * P ortla n d, O re g on's , W illa m ette riv erfro nt d ev elo pm en t h as e x plo ded w ith se v en c o ndom in iu m to w ers a n d a n ew h ealth s c ie n ces c en te r u nder c o nstr u ctio n. T he h ealth s c ie n ce c o m ple x is to b e lin ked w ith O re g on H ealth S cie n ces U niv ers ity ( O HSU ), w hic h i s h ig h o n a n earb y h ill, w ith a n a eria l c ab le t r a m . T he aeria l tr a m lin kin g th e w ate rfro nt dis tr ic t to O HSU is to su pport th e univ ers ity ex pan sio n, to in cre ase b io te ch nolo gy re se arc h , a n d to b eco m e P ortl a n d's ic o n e q uiv ale n t to S eattle 's S pace N eed le . A ll o f th e h ype tu rn ed s o uth w hen n ew s f ro m a h earin g s u ggeste d th at th e r e al b u dget f o r th e tr a m c o nstr u ctio n, o rig in ally e stim ate d a t $ 15 m illio n, is g oin g to b e a b out $ 55–$60 m illio n, n early tr ip le th e o rig in al e stim ate . T he e stim ate c o uld e v en g o h ig her.

C om mis sio ners w an t t o f in d o ut w hy c ity s ta ff k now in gly r e lie d o n f la w ed e stim ate s. M ik e L in dberg , p re sid en t o f t h e n onpro fit A eria l T ra n sp orta tio n I n c., a ck now le d ged “ th e $ 15 m il lio n n um ber w as n ot a g ood n um ber. I t w as s im ply a g uesstim ate .” C om mis sio ner E rik S te n s a id , “ T hose n um bers w ere p re se n te d a s m uch m ore f ir m th an th ey a p pear to h av e b een … . I t a p pears t h e a ctu al d esig n w asn 't c o ste d o ut. T hat's p re tty s h oddy.” © C lif f G ra y * T he O re g onia n, J a n uary 1 3, 2 006, b y F ra n k R yan , p ag es A 1 a n d A 14, a n d A pril 2 , 2 006, p ag e A 1. M eth ods f o r E stim atin g P ro je ct T im es a n d C osts T op -D ow n A ppro ach es f o r E stim atin g P ro je ct T im es a n d C osts A t th e s tr a te g ic le v el to p-d ow n e stim atin g m eth ods a re u se d to e v alu ate th e p ro je ct p ro posa l. S om etim es m uch o f th e in fo rm atio n n eed ed to d eriv e a cc u ra te tim e a n d c o st e stim ate s is n ot a v aila b le in th e in itia l p hase o f th e 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 0/4 4 p ro je ct— fo r e x am ple , d esig n is n ot f in aliz ed . I n th ese s itu atio ns to p-d ow n e stim ate s a re u se d u ntil th e ta sk s in t h e W BS a re c le arly d efin ed .

C on se n su s M eth ods T his m eth od s im ply u se s th e p oole d e x perie n ce o f s e n io r a n d/o r m id dle m an ag ers to e stim ate th e to ta l p ro je ct d ura tio n an d co st. T his ty pic ally in vo lv es a m eetin g w here ex perts d is c u ss, arg ue, an d u ltim ate ly re ach a d ecis io n a s to th eir b est g uess e stim ate . F ir m s s e ek in g g re ate r r ig or w ill u se th e D elp hi M eth od to m ak e th ese m acro e stim ate s. S ee S nap sh ot f ro m P ra ctic e: T he D elp hi M eth od. I t i s i m porta n t t o r e co gniz e t h at t h ese f ir s t t o p-d ow n e stim ate s a re o nly a r o ugh c u t a n d t y pic ally o ccu r i n t h e “ co ncep tu al” s ta g e o f t h e p ro je ct. T he t o p-d ow n e stim ate s a re h elp fu l i n i n itia l d ev elo pm en t o f a c o m ple te p la n .

H ow ev er, su ch estim ate s are so m etim es sig nif ic an tly off th e m ark becau se little deta ile d in fo rm atio n is g ath ere d . A t th is le v el in div id ual w ork ite m s a re n ot id en tif ie d . O r, in a f e w c a se s, th e to p-d ow n e stim ate s a re n ot re alis ti c b ecau se to p m an ag em en t “ w an ts th e p ro je ct.” N ev erth ele ss, th e in itia l to p-d ow n e stim ate s a re h elp fu l in d ete rm in in g w heth er th e p ro je ct w arra n ts m ore f o rm al p la n nin g, w hic h w ould in clu de m ore d eta ile d e stim ate s. B e c are fu l th at m acro e stim ate s m ad e b y s e n io r m an ag ers a re n ot d ic ta te d to lo w er le v el m an ag ers w ho m ig ht f e el c o m pelle d t o a ccep t t h e e stim ate s e v en i f t h ey b elie v e r e so urc es a re i n ad eq uate . 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 1 /4 4 P ag e 1 36 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . S N APSH OT F R O M P R A CTIC E T he D elp hi M eth od O rig in ally d ev elo ped b y t h e R A ND C orp ora ti o n i n 1 969 f o r t e ch nolo gic al f o re castin g, t h e D elp hi M eth od i s a g ro up d ecis io n p ro cess a b out th e lik elih ood th at c erta in e v en ts w ill o ccu r. T he D elp hi M eth od m ak es u se o f a p an el o f e x perts fa m ilia r w ith th e k in d o f p ro je ct in q uestio n. T he n otio n is th at w ell- in fo rm ed in div id uals , c allin g o n th eir i n sig hts a n d e x perie n ce, a re b ette r e q uip ped to e stim ate p ro je ct c o sts /tim es th an th eo re tic al a p pro ach es o r s ta tis tic al m eth ods. T heir re sp onse s to estim ate q uestio nnair e s are an onym ou s, an d th ey are p ro vid ed w ith a su m mary o f o pin io ns. E xperts a re th en e n co ura g ed to re co nsid er, a n d if a p pro pria te , to c h an ge th eir p re v io us e stim ate in lig ht o f th e r e p lie s o f o th er e x perts . A fte r tw o o r th re e ro unds it is b elie v ed th at th e g ro up w ill c o nverg e to w ard th e “ b est” r e sp onse th ro ugh th is c o nse n su s p ro cess. T he m id poin t o f r e sp onse s is s ta tis tic ally c a te g oriz ed b y th e m ed ia n s c o re .

I n e ac h s u cceed in g r o u nd o f q uestio nnair e s, t h e r a n ge o f r e sp onse s b y t h e p an elis ts w ill p re su m ab ly d ecre ase a n d t h e m ed ia n w ill m ove t o w ard w hat i s d eem ed t o b e t h e “ co rre ct” e stim ate . O ne d is tin ct a d van ta g e o f th e D elp hi M eth od is th at th e e x perts n ev er n eed to b e b ro ught to geth er p hysic ally . T he p ro cess a ls o d oes n ot re q uir e c o m ple te a g re e m en t b y a ll p an elis ts , s in ce th e m ajo rity o pin io n is re p re se n te d b y th e m ed ia n . S in ce th e re sp onse s a re a n onym ous, th e p itf a lls o f e g o, d o m in eerin g p ers o nalitie s, a n d th e “ b an dw ag on o r h alo e ff e ct” in r e sp onse s a re a ll a v oid ed . O n th e o th er h an d, f u tu re d ev elo pm en ts a re n o t a lw ay s p re d ic te d c o rre ctly b y i te ra tiv e c o nse n su s n or b y e x perts , b ut a t t i m es b y c re ativ e, “ o ff t h e w all” t h in kin g.

A lth ough y our a u th ors p re fe r to a v oid th e to p-d ow n a p pro ach if p ossib le , w e h av e w itn esse d su rp ris in g a ccu ra cy in e stim atin g p ro je ct d ura tio n a n d c o st in is o la te d c ase s. S om e e x am ple s a re b uild in g a m an ufa ctu rin g p la n t, b uild in g a d is tr ib utio n w are h ouse , d ev elo pin g a ir c o ntr o l f o r s k ysc ra p er b uild in gs, a n d r o ad c o nstr u ctio n.

H ow ev er, w e h av e a ls o w itn esse d s o m e h orre n dous m is c alc u la tio ns, u su ally in a re as w here th e te ch nolo gy is n ew a n d u npro ven . T op-d ow n m eth ods c an b e u se fu l i f e x perie n ce a n d j u dgm en t h av e b een a ccu ra te i n t h e p ast.

R atio M eth ods T op-d ow n m eth ods ( s o m etim es c alle d p ara m etr ic ) u su ally u se r a tio s, o r s u rro gate s, to e stim ate p ro je ct tim es o r c o sts . T op-d ow n a p pro ach es a re o fte n u se d in th e c o ncep t o r “ n eed ” p hase o f a p ro je ct to g et a n in itia l d ura tio n a n d c o st e stim ate f o r th e p ro je ct. F or e x am ple , c o ntr a cto rs f re q uen tly u se n um ber o f s q uare f e et to e stim ate th e c o st a n d t im e t o b uild a h ouse ; t h at i s , a h ouse o f 2 ,7 00 s q uare f e et m ig ht c o st $ 160 p er s q uare f o ot ( 2 ,7 00 f e et × $ 160 p er fo ot e q uals $ 432,0 00). L ik ew is e , k now in g th e sq uare fe et a n d d olla rs p er sq uare fo ot, e x perie n ce s u ggests it s h ould ta k e a p pro xim ate ly 1 00 d ay s to c o m ple te . T w o o th er c o m mon e x am ple s o f to p-d ow n c o st e stim ate s a re t h e c o st f o r a n ew p la n t e stim ate d b y c ap acity s iz e, o r a s o ftw are p ro duct e stim ate d b y f e atu re s a n d c o m ple x ity . 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 2/4 4 P ag e 1 37 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . F IG URE 5 .1 A pportio n M eth od o f A llo ca tin g P ro je ct C osts U sin g t h e W ork B re a k dow n S tr u ctu re A pportio n M eth ods T his m eth od is a n e x te n sio n to th e ra tio m eth od. A pportio n m en t is u se d w hen p ro je cts c lo se ly fo llo w p ast p ro je cts in f e atu re s a n d c o sts . G iv en g o od h is to ric al d ata , e stim ate s c an b e m ad e q uic k ly w ith little e ff o rt a n d r e aso nab le a ccu ra cy. T his m eth od is v ery c o m mon in p ro je cts th at a re r e la tiv ely s ta n dard b ut h av e s o m e s m all v aria tio n o r c u sto m iz atio n. A nyone w ho h as b orro w ed m oney f ro m a b an k to b uild a h ouse h as b een e x pose d to th is p ro cess. G iv en a n e stim ate d to ta l co st fo r th e h ouse , b an ks an d th e F H A (F ed era l H ousin g A uth ority ) au th oriz e p ay to th e c o ntr a cto r b y c o m ple tio n o f s p ecif ic s e g m en ts o f th e h ouse . F or e x am ple , f o undatio n m ig ht r e p re se n t 3 p erc en t o f t h e t o ta l l o an , f ra m in g 2 5 p erc en t, e le ctr ic , p lu m bin g a n d h eatin g 1 5 p erc en t, e tc . P ay m en ts a re m ad e a s t h ese i te m s a re c o m ple te d . A n a n alo gous p ro cess is u se d b y s o m e c o m pan ie s th at a p portio n c o sts to d eliv era b le s in t h e W BS— giv en a v era g e c o st p erc en ta g es fro m p ast p ro je cts . F ig ure 5 .1 p re se n ts a n e x am ple s im ila r to o ne f o und in p ra ctic e. A ssu m in g th e to ta l p ro je ct c o st is e stim ate d , u sin g a to p-d ow n e stim ate , to b e $ 500,0 00, th e c o sts a re a p portio ned a s a p erc en ta g e o f th e to ta l c o st. F or e x am ple , th e c o sts a p portio ned to th e “ D ocu m en t” d eliv era b le a re 5 p erc en t o f th e to ta l, o r $ 25,0 00. T he s u bdeliv era b le s “ D oc-1 a n d D oc-2 ” a re a llo cate d 2 a n d 3 p erc en t o f t h e t o ta l— $10,0 00 a n d $ 15,0 00, r e sp ectiv ely .

F unctio n P oin t M eth ods f o r S oftw are a n d S yste m P ro je c ts I n th e so ftw are in dustr y , so ftw are dev elo pm en t pro je cts are fre q uen tly estim ate d usin g w eig hte d m acro v aria b le s c alle d “ fu nctio n p oin ts ” o r m ajo r p ara m ete rs s u ch a s n um ber o f i n puts , n um ber o f o utp uts , n um ber o f i n quir ie s, num ber of data file s, an d num ber of in te rfa ces. T hese w eig hte d varia b le s are ad ju ste d fo r a c o m ple x ity f a cto r a n d a d ded . T he t o ta l a d ju ste d c o unt p ro vid es t h e b asis f o r e stim atin g t h e l a b or e ff o rt a n d c o st f o r a p ro je ct ( u su ally u sin g a r e g re ssio n f o rm ula d eriv ed f ro m d ata o f p ast p ro je cts ). T his l a tte r m eth od a ssu m es a d eq uate h is to ric al d ata b y ty pe o f s o ftw are p ro je ct fo r th e in dustr y — fo r e x am ple , M IS s y ste m s. In th e U .S .

s o ftw are in dustr y , o ne-p ers o n m onth r e p re se n ts o n a v era g e f iv e f u nctio n p oin ts . A p ers o n w ork in g o ne m onth c an g en era te o n a v era g e (a cro ss a ll ty pes o f so ftw are p ro je cts ) a b out fiv e fu nctio n p oin ts . O f c o urs e e ach o rg an iz atio n n eed s t o d ev elo p i ts o w n a v era g e f o r i ts s p ecif ic t y pe o f w ork . S uch h is to ric al d ata p ro vid e a b asis 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 3/4 4 f o r e stim atin g th e p ro je ct d ura tio n. V aria tio ns o f th is to p-d ow n a p pro ach a re u se d b y c o m pan ie s s u ch a s I B M , B an k o f A m eric a, S ears R oeb uck , H P, A T& T, F ord M oto rs , G E, D uP ont a n d m an y o th ers . S ee T ab le 5 .2 a n d T ab le 5 .3 f o r a s im plif ie d e x am ple o f f u nctio n p oin t c o unt m eth odolo gy. 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 4/4 4 P ag e 1 38 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . T A BLE 5 .2 S im plif ie d B asic F unctio n P oin t C ou nt P ro cess f o r a P ro sp ec tiv e P ro je ct o r D eliv er a b le F ro m h is to ric al d ata th e o rg an iz atio n d ev elo ped th e w eig htin g s c h em e fo r c o m ple x ity fo und in T ab le 5 .2 . F unctio n p oin ts a re d eriv ed f ro m m ultip ly in g t h e n um ber o f k in ds o f e le m en ts b y w eig hte d c o m ple x ity . T ab le 5 .3 sh o w s th e d ata c o lle cte d fo r a sp ecif ic ta sk o r d eliv era b le : P atie n t A dm ittin g a n d B illin g— th e n um ber o f i n puts , o utp uts , i n quir ie s, f ile s, a n d i n te rfa ces a lo ng w ith t h e e x pecte d c o m ple x ity r a tin g. F in ally , t h e a p plic atio n o f th e e le m en t c o unt is a p plie d a n d th e f u nctio n p oin t c o unt to ta l is 6 60. G iv en th is c o unt a n d th e f a ct th at o ne-p ers o n m onth h as h is to ric ally b een e q ual to 5 fu nctio n p oin ts , th e jo b w ill re q uir e 1 32 p ers o n m onth s ( 6 6 0/5 = 1 32). A ssu m in g y ou h av e 1 0 p ro gra m mers w ho c an w ork o n th is ta sk , th e d ura tio n w ould b e a p pro xim ate ly 1 3 m onth s. T he c o st is e asily d eriv ed b y m ultip ly in g th e la b or r a te p er m onth tim es 1 32 p ers o n m onth s. F or e x am ple , i f t h e m onth ly p ro gra m mer 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 5/4 4 P ag e 1 39 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . r a te is $ 4,0 00, th en th e e stim ate d c o st w ould b e $ 528,0 00 (1 32 × 4 ,0 00). A lth ough fu nctio n p oin t m etr ic s a re u se fu l, t h eir a ccu ra cy d ep en ds o n a d eq uate h is to ric al d ata , c u rr e n cy o f d ata , a n d r e le v an cy o f t h e p ro je ct/d eliv era b le t o p ast a v era g es.

T A BLE 5 .3 E xam ple : F unctio n P oin t C ou nt M eth od L ea rn in g C urv es S om e p ro je cts r e q uir e th at th e s a m e ta sk , g ro up o f ta sk s, o r p ro duct b e r e p eate d s e v era l tim es. M an ag ers k now i n tu itiv ely t h at t h e t im e t o p erfo rm a t a sk i m pro ves w ith r e p etitio n. T his p hen om en on i s e sp ecia lly t r u e o f t a sk s t h at a re la b or in te n siv e. I n th ese c ir c u m sta n ces th e p atte rn o f im pro vem en t p hen om en on c an b e u se d to p re d ic t t h e re d uctio n in tim e to p erfo rm th e ta sk . F ro m e m pir ic al e v id en ce a cro ss a ll in dustr ie s, th e p atte rn o f th is i m pro vem en t h as b een q uan tif ie d in th e l e a rn in g c u rv e ( a ls o k now n a s im pro vem en t c u rv e, e x perie n ce c u rv e, a n d i n dustr ia l p ro gre ss c u rv e), w hic h i s d esc rib ed b y t h e f o llo w in g r e la tio nsh ip : E ach t im e t h e o utp ut q uantity d ouble s, t h e u nit l a bor h ours a re r e d uced a t a c o nsta nt r a te . I n p ra ctic e th e im pro vem en t ra tio m ay v ary fro m 6 0 p erc en t, re p re se n tin g v ery la rg e im pro vem en t, to 1 00 p erc en t, re p re se n tin g n o im pro vem en t a t a ll. G en era lly , a s th e d if f ic u lty o f th e w ork d ecre ase s th e e x pecte d i m pro vem en t a ls o d ecre ase s a n d th e im pro vem en t r a tio th at is u se d b eco m es g re ate r. O ne s ig nif ic an t f a cto r to c o nsid er i s t h e p ro portio n o f l a b or i n t h e t a sk i n r e la tio n t o m ach in e-p aced w ork . O bvio usly , a l o w er p erc en ta g e o f im pro vem en t c an o ccu r o nly in o pera tio ns w ith h ig h la b or c o nte n t. A ppen dix 5 .1 a t th e e n d o f th e c h ap te r p ro vid es a d eta ile d e x am ple o f h ow th e im pro vem en t p hen om en on c an b e u se d to e stim ate tim e a n d c o st fo r r e p etitiv e t a sk s. T he m ain d is a d van ta g e o f to p-d ow n a p pro ach es to e stim atin g is s im ply th at th e tim e a n d c o st f o r a s p ecif ic t a sk a re n ot c o nsid ere d . G ro upin g m an y ta sk s in to a c o m mon b ask et e n co ura g es e rro rs o f o m is sio n a n d th e u se o f i m pose d t im es a n d c o sts . M ic ro e stim atin g m eth ods a re u su ally m ore a ccu ra te t h an m acro m eth ods. B otto m -U p A ppro ach es f o r E stim atin g P ro je ct T im es a n d C osts T em pla te M eth ods I f th e p ro je ct is s im ila r to p ast p ro je cts , th e c o sts f ro m p ast p ro je cts c an b e u se d a s a s ta rtin g p oin t f o r th e n ew p ro je ct. D if f e re n ces in th e new pro je ct can be note d an d past tim es an d co sts ad ju ste d to re fle ct th ese 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 6/4 4 d if f e re n ces. F or e x am ple , a s h ip re p air d ry dock fir m h as a s e t o f s ta n dard re p air p ro je cts (i.e ., te m pla te s fo r o verh au l, e le ctr ic al, m ech an ic al) t h at a re u se d a s s ta rtin g p oin ts f o r e stim atin g t h e c o st a n d d ura tio n o f a n y n ew p ro je ct. D if f e re n ces fro m th e a p pro pria te s ta n dard iz ed p ro je ct a re n ote d (fo r tim es, c o sts , a n d re so urc es) a n d c h an ges a re m ad e. T his a p pro ach e n ab le s t h e f ir m t o d ev elo p a p ote n tia l s c h ed ule , e stim ate c o sts , a n d d ev elo p a b udget in a v ery s h ort tim e s p an . D ev elo pm en t o f s u ch te m pla te s in a d ata b ase c an q uic k ly re d uce e stim ate e rro rs .

P ara m etr ic P ro ced u re s A pplie d t o S pecif ic T ask s J u st a s p ara m etr ic te ch niq ues s u ch a s c o st p er s q uare fo ot c an b e th e s o urc e o f to p-d ow n e stim ate s, th e s a m e t e ch niq ue c an b e a p plie d t o s p ecif ic t a sk s. F or e x am ple , a s p art o f a n M S O ff ic e c o nvers io n p ro je ct, 3 6 d if f e re n t c o m pute r w ork sta tio ns n eed ed t o b e c o nverte d . B ase d o n p ast c o nvers io n p ro je cts , t h e p ro je ct 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 7/4 4 P ag e 1 40 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . m an ag er dete rm in ed th at on av era g e one pers o n co uld co nvert th re e w ork sta tio ns per day.

T here fo re th e ta sk o f c o nvertin g th e 3 6 w ork sta tio ns w ould ta k e th re e te ch nic ia n s fo ur d ay s [ (3 6/3 )/3 ]. S im ila rly , to e stim ate th e w allp ap erin g a llo w an ce o n a h ouse r e m odel, th e c o ntr a cto r f ig ure d a c o st o f $ 5 p er sq uare y ard o f w allp ap er a n d $ 2 p er y ard to in sta ll it, fo r a to ta l c o st o f $ 7. B y m easu rin g th e le n gth a n d h eig ht o f a ll th e w alls s h e w as a b le to c alc u la te th e to ta l a re a in s q uare y ard s a n d m ultip ly i t b y $ 7.

R an ge E stim atin g W hen d o yo u u se ra nge estim atin g? R an ge e stim atin g w ork s b est w hen w ork p ack ag es h av e sig nif ic an t u ncerta in ty asso cia te d w ith th e tim e o r co st to co m ple te . If th e w ork p ack ag e is ro utin e an d carrie s little u ncerta in ty , u sin g a p ers o n m ost f a m ilia r w ith th e w ork p ack ag e is u su ally th e b est a p pro ach . T hey k now f ro m e x perie n ce o r k now w here to fin d th e in fo rm atio n to e stim ate w ork p ack ag e d ura tio ns a n d c o sts . H ow ev er, w hen w ork p ack ag es h av e s ig nif ic an t u ncerta in ty a sso cia te d w ith th e tim e o r c o st to c o m ple te , it is a p ru den t p olic y to re q uir e th re e tim e e stim ate s— lo w , a v era g e, a n d h ig h (b orro w ed o ff o f P E RT m eth odolo gy th at u se s p ro bab ility d is tr ib utio ns). T he l o w t o h ig h g iv e a r a n ge w ith in w hic h t h e a v era g e e stim ate w ill f a ll. D ete rm in in g t h e lo w a n d h ig h e stim ate s fo r th e a ctiv ity is in flu en ced b y fa cto rs s u ch a s c o m ple x ity , te ch nolo gy, n ew ness, f a m ilia rity . H ow d o y o u g et th e e stim ate s? S in ce ra n ge e stim atin g w ork s b est fo r w ork p ack ag es th at h av e s ig nif ic an t u ncerta in ty , h av in g a g ro up d ete rm in e th e lo w , a v era g e, a n d h ig h c o st o r d ura tio n g iv es b est re su lts . G ro up e stim atin g te n ds to r e fin e e x tr e m es b y b rin gin g m ore e v alu ativ e ju dgm en ts to th e e stim ate a n d p ote n tia l r is k s.

T he ju dgm en t o f o th ers in a g ro up h elp s to m odera te e x tr e m e p erc eiv ed ris k s a sso cia te d w ith a tim e o r c o st e stim ate . I n volv in g o th ers i n m ak in g a ctiv ity e stim ate s g ain s b uy i n a n d c re d ib ility t o t h e e stim ate . F ig ure 5 .2 p re se n ts a n a b rid ged e stim atin g te m pla te u sin g th re e tim e e stim ate s f o r w ork p ack ag es d ev elo ped b y a c ro ss f u nctio nal g ro up(s ) o f p ro je ct s ta k eh old ers . T he g ro up e stim ate s s h ow th e lo w , a v era g e, a n d h ig h f o r e ach w ork 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 8/4 4 P ag e 1 41 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . p ack ag e. T he R is k L ev el co lu m n is th e gro up's in dep en den t asse ssm en t of th e deg re e of c o nfid en ce th at th e actu al tim e w ill b e v ery clo se to th e estim ate . In a se n se th is n um ber r e p re se n ts th e gro up's ev alu atio n of m an y fa cto rs (e .g ., co m ple x ity , te ch nolo gy) th at m ig ht i m pact th e a v era g e tim e e stim ate . I n o ur e x am ple , th e g ro up f e els w ork p ack ag es 1 04, 1 08, 1 1 0, 1 11 , a n d 1 1 4 h av e a h ig h c h an ce th at th e a v era g e tim e m ay v ary fro m e x pecte d . L ik ew is e , th e g ro up's c o nfid en ce fe els th e r is k o f w ork p ack ag es 1 02, 1 05 a n d 1 1 2 n ot m ate ria liz in g a s e x pecte d i s l o w .

F IG URE 5 .2 R an ge E stim atin g T em pla te H ow d o y o u u se t h e e stim ate ? G ro up r a n ge e stim atin g g iv es t h e p ro je ct m an ag er a n d o w ner a n o pportu nity t o a sse ss th e c o nfid en ce a sso cia te d w ith p ro je ct tim es (a n d/o r c o sts ). F or e x am ple , a c o ntr a cto r re sp onsib le fo r b uild in g a h ig h r is e a p artm en t b uild in g c an te ll th e o w ner th at th e p ro je ct w ill c o st b etw een 3 .5 a n d 4 .1 m illio n d olla rs a n d t a k e b etw een s ix a n d n in e m onth s t o c o m ple te . T he a p pro ach h elp s t o r e d uce s u rp ris e s a s t h e p ro je ct p ro gre sse s. T he ra n ge e stim atin g m eth od a ls o p ro vid es a b asis fo r a sse ssin g ris k , m an ag in g re so urc es, a n d d ete rm in in g th e p ro je ct co ntin gen cy fu nd. (S ee C hap te r 7 fo r a dis c u ssio n of co ntin gen cy fu nds.) R an ge e stim atin g is p opula r in s o ftw are a n d n ew p ro duct p ro je cts w here u p-fro nt r e q uir e m en ts a re f u zzy a n d n ot w ell k now n. G ro up r a n ge e stim atin g i s o fte n u se d w ith p hase e stim atin g, w hic h i s d is c u sse d n ex t.

A H yb rid : P hase E stim atin g T his a p pro ach b eg in s w ith a to p-d ow n e stim ate fo r th e p ro je ct a n d th en re fin es e stim ate s fo r p hase s o f th e p ro je ct as it is im ple m en te d . S om e p ro je cts b y th eir n atu re can not b e rig oro usly d efin ed b ecau se o f th e u ncerta in ty o f d esig n o r th e f in al p ro duct. A lth ough r a re , s u ch p ro je cts d o e x is t. T hese p ro je cts a re o fte n f o und i n ae ro sp ace pro je cts , IT pro je cts , new te ch nolo gy pro je cts , an d co nstr u ctio n pro je cts w here desig n is i n co m ple te . I n t h ese p ro je cts , p hase o r l if e -c y cle e stim atin g i s f re q uen tly u se d . P hase e stim atin g is u se d w hen a n u nusu al a m ount o f u ncerta in ty s u rro unds a p ro je ct a n d it is im pra ctic al to e stim ate tim es a n d c o sts f o r th e e n tir e p ro je ct. P hase e stim atin g u se s a tw o-e sti m ate s y ste m o ver th e lif e o f th e p ro je ct. A d eta ile d estim ate is d ev elo ped fo r th e im med ia te p hase an d a m acro estim ate is m ad e fo r th e r e m ain in g p hase s o f t h e p ro je ct. F ig ure 5 .3 d ep ic ts t h e p hase s o f a p ro je ct a n d t h e p ro gre ssio n o f e stim ate s o ver i ts l if e .

F IG URE 5 .3 P hase E stim atin g o ver P ro je ct L if e C ycle 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 1 9/4 4 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 0/4 4 P ag e 1 42 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . S N APSH OT F R O M P R A CTIC E E stim ate A ccu ra cy T he s m alle r th e e le m en t o f a w ork p ack ag e, th e m ore a c cu ra te th e o v era ll e stim ate is li k ely to b e. T he e x te n t o f th is i m pro vem en t v arie s b y ty pe o f p ro je ct. T he ta b le b elo w is d ev elo ped to re fle ct th is o bse rv atio n. F or ex am ple , i n fo rm atio n te ch nolo gy p ro je cts th at d ete rm in e th eir tim e a n d c o st e stim ate s in th e c o ncep tu al s ta g e c an e x pect th eir “ actu als ” to e rr u p to 2 00 p erc en t o ver c o st a n d d u ra tio n a n d, p erh ap s, a s m uch a s 3 0 p erc en t u nder e stim ate s.

C onvers e ly , e stim ate s fo r b uild in gs, ro ad s, e tc ., m ad e a fte r th e w ork p ack ag es a re c le arly d efin ed , h av e a s m alle r e rro r in a ctu al c o sts a n d tim es o f 1 5 p erc en t o ver e stim ate a n d 5 p erc en t le ss th an e stim ate . A lth ough th ese e stim ate s v ary b y p ro je ct, th ey c an se rv e a s b allp ark n u m bers fo r p ro je ct sta k eh old ers se le ctin g h ow p ro je ct tim e a n d c o st e stim ate s w ill b e d eriv ed .

F or e x am ple , w hen th e p ro je ct n eed is d ete rm in ed , a m acro e stim ate o f th e p ro je ct c o st a n d d ura tio n is m ad e s o an aly sis an d decis io ns can be m ad e. S im ulta n eo usly a deta ile d estim ate is m ad e fo r deriv in g pro je ct s p ecif ic atio ns an d a m acro estim ate fo r th e re m ain der of th e pro je ct. A s th e pro je ct pro gre sse s an d s p ecif ic atio ns a re s o lid if ie d , a d eta ile d e stim ate fo r d esig n is m ad e a n d a m acro e stim ate fo r th e re m ain der o f t h e p ro je ct is co m pute d . C le arly , as th e p ro je ct p ro gre sse s th ro ugh its lif e cy cle an d m ore in fo rm atio n is a v aila b le , th e re lia b ility of th e estim ate s sh ould be im pro vin g. See Snap sh ot fro m Pra ctic e: E stim atin g A ccu ra cy. P hase e stim atin g is p re fe rre d b y th ose w ork in g o n p ro je cts w here th e fin al p ro duct is n ot k now n a n d th e u ncerta in ty is v ery la rg e— fo r e x am ple , th e in te g ra tio n o f w ir e le ss p hones a n d c o m pute rs . T he c o m mitm en t to c o st a n d s c h ed ule is o nly n ecessa ry o ver th e n ex t p hase o f th e p ro je ct a n d c o m mitm en t to u nre alis tic fu tu re s c h ed ule s a n d c o sts b ase d o n p oor in fo rm atio n is a v oid ed . T his p ro gre ssiv e m acro /m ic ro m eth od p ro vid es a s tr o nger b asis f o r u sin g s c h ed ule a n d c o st e stim ate s t o m an ag e p ro gre ss d urin g t h e n ex t p hase . U nfo rtu nate ly y our c u sto m er— in te rn al o r e x te rn al— will w an t a n a ccu ra te e stim ate o f s c h ed ule a n d c o st th e m om en t t h e d ecis io n i s m ad e t o i m ple m en t t h e p ro je ct. A dditio nally , t h e c u sto m er w ho i s p ay in g f o r t h e p ro je ct o fte n p erc eiv es p hase e stim atin g a s a b la n k c h eck b ecau se c o sts a n d s c h ed ule s a re n ot fir m o ver m ost o f th e p ro je ct lif e c y cle . E ven th ough th e r e aso ns f o r p hase e stim atin g a re s o und a n d le g itim ate , m ost c u sto m ers h av e t o b e so ld o n its le g itim acy. A m ajo r a d van ta g e fo r th e c u sto m er is th e o pportu nity to c h an ge fe atu re s, re - e v alu ate , o r e v en c an cel t h e p ro je ct i n e ach n ew p hase . I n c o nclu sio n, p hase e stim atin g i s v ery u se fu l i n p ro je cts t h at p osse ss h uge u ncerta in tie s c o ncern in g t h e f in al n atu re ( s h ap e, s iz e, f e atu re s) o f t h e p ro je ct. S ee F ig ure 5 .4 f o r a s u m mary o f t h e d if f e re n ces b etw een t o p-d ow n a n d b otto m -u p e stim ate s. 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 1/4 4 P ag e 1 43 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . F IG URE 5 .4 T op -D ow n a n d B otto m -U p E stim ate s O bta in in g a ccu ra te e stim ate s i s a c h alle n ge. C om mitte d o rg an iz atio ns a ccep t t h e c h alle n ge o f c o m in g u p w ith m ean in gfu l estim ate s an d in vest h eav ily in d ev elo pin g th eir cap acity to d o so . A ccu ra te estim ate s re d uce u ncerta in ty a n d s u pport a d is c ip lin e f o r e ff e ctiv ely m an ag in g p ro je cts .

L ev el o f D eta il L ev el o f d eta il is d if f e re n t f o r d if f e re n t le v els o f m an ag em en t. A t a n y le v el th e d eta il s h ould b e n o m ore th an is n ecessa ry a n d su ff ic ie n t. T op m an ag em en t in te re sts u su ally c en te r o n th e to ta l p ro je ct a n d m ajo r m ile sto ne e v en ts t h at m ark m ajo r a cco m plis h m en ts — e.g ., “ B uild O il P la tf o rm i n t h e N orth S ea” o r “ C om ple te P ro to ty pe.” M id dle m an ag em en t m ig ht c en te r o n o ne s e g m en t o f t h e p ro je ct o r o ne m ile sto ne. F ir s t- lin e m an ag ers ' i n te re sts m ay b e lim ite d to o n e ta sk o r w ork p ack ag e. O ne o f th e b eau tie s o f W BS is th e a b ility to a g gre g ate n etw ork i n fo rm atio n s o e ach l e v el o f m an ag em en t c an h av e t h e k in d o f i n fo rm atio n n ecessa ry t o m ak e d ecis io ns. G ettin g th e le v el o f d eta il in th e W BS to m atc h m an ag em en t n eed s fo r e ff e ctiv e im ple m en ta tio n is c ru cia l, b ut th e d elic ate b ala n ce is d if f ic u lt to f in d. S ee S nap sh ot f ro m P ra ctic e: L ev el o f D eta il. T he le v el o f d eta il in t h e W BS v arie s w it h th e c o m ple x ity o f th e p ro je ct; th e n eed fo r c o ntr o l; th e p ro je ct s iz e, c o st, d ura tio n; a n d o th er fa cto rs . If th e str u ctu re re fle cts ex cessiv e deta il, th ere is a te n den cy to bre ak th e w ork eff o rt in to d ep artm en t assig nm en ts . T his te n den cy can beco m e a barrie r to su ccess, sin ce th e em phasis w ill be on d ep artm en ta l o utc o m es ra th er th an o n d eliv era b le o utc o m es. E xcessiv e d eta il a ls o m ean s m ore u npro ductiv e p ap erw ork . N ote th at if th e le v el o f th e W BS is in cre ase d b y o ne, th e n um ber o f c o st a cco unts m ay in cre ase g eo m etr ic ally . O n th e o th er h an d, if th e le v el o f d eta il is n ot ad eq uate , an o rg an iz atio n u nit m ay fin d th e s tr u ctu re f a lls s h ort o f m eetin g i ts n eed s. F ortu nate ly , t h e W BS h as b uilt- in f le x ib ility . P artic ip atin g o rg an iz atio n u nits m ay ex pan d th eir p ortio n o f th e str u ctu re to m eet th eir sp ecia l n eed s. F or ex am ple , th e en gin eerin g d ep artm en t m ay w is h to f u rth er b re ak th eir w ork o n a d eliv era b le in to s m alle r p ack ag es b y e le ctr ic al, c iv il, a n d 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 2/4 4 m ech an ic al. S im ila rly , t h e m ark etin g d ep artm en t m ay w is h t o b re ak t h eir n ew p ro duct p ro m otio n i n to T V, r a d io , p erio dic als , a n d n ew sp ap ers . 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 3/4 4 P ag e 1 44 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . S N APSH OT F R O M P R A CTIC E L ev el o f D eta il— Rule o f T hum b P ra ctic in g p ro je ct m an ag ers a d vocate k eep in g t h e l e v el o f d eta il t o a m in im um . B ut t h ere a re l im its t o t h is s u ggestio n.

O ne o f t h e m ost f re q uen t e rro rs o f n ew p ro je ct m an ag ers i s t o f o rg et t h at t h e t a sk t im e e stim ate w ill b e u se d t o c o ntr o l s c h ed ule an d co st p erfo rm an ce. A fre q uen t ru le o f th um b u se d b y p ra ctic in g p ro je ct m an ag ers sa y s th at a ta sk d ura tio n sh o uld n ot e x ceed 5 w ork day s o r a t th e m ost 1 0 w ork day s, if w ork day s a re th e tim e u nits u se d fo r th e p ro je ct. S uch a r u le p ro bab ly w ill r e su lt in a m ore d eta ile d n etw ork , b ut th e a d ditio nal d eta il p ay s o ff in c o ntr o llin g s c h ed ule a n d c o st a s t h e p ro je ct p ro gre sse s. S uppose t h e t a sk i s “ b uild p ro to ty pe c o m pute r- c o ntr o lle d c o nvey or b elt,” t h e t im e e stim ate i s 4 0 w ork day s, a n d t h e b udget $ 30 0,0 00. I t m ay b e b ette r to d iv id e th e ta sk in to s e v en o r e ig ht s m alle r ta sk s f o r c o ntr o l p urp ose s. I f o ne o f t h e s m alle r t a sk s g ets b eh in d b ecau se o f p ro ble m s o r a p oor t im e e stim ate , i t w ill b e p ossib le t o t a k e c o rre ctiv e a ctio n q uic k ly a n d a v oid d ela y in g s u ccessiv e ta sk s a n d th e p ro je ct. I f th e s in gle ta sk o f 4 0 w ork day s is u se d , it is p ossib le t h at n o c o rre ctiv e a ctio n w ould b e ta k en u nti l d ay 4 0 , s in ce m an y p eo ple h av e a te n den cy to “ w ait a n d s e e” o r a v oid a d m ittin g t h ey a re b eh in d o r p assin g o n b ad n ew s; t h e r e su lt m ay m ea n f a r m ore t h an 5 d ay s b eh in d s c h ed ule . T he 5 -to 1 0-d ay r u le o f th um b a p plie s to c o st a n d p erfo rm an ce g o als . I f u sin g th e r u le o f th um b s u ggeste d a b ove r e su lts i n t o o m an y n etw ork t a sk s, a n a lte rn ati v e i s a v aila b le , b u t i t h as c o nditio ns. T he a c tiv ity t im e c an b e e x te n ded b ey ond th e 5 -to 1 0-d ay r u le o nly I F c o ntr o l m onito rin g c h eck poin ts fo r s e g m en ts o f th e ta sk c a n b e e sta b lis h ed s o c le ar m easu re s o f p ro gre ss c an b e i d en tif ie d b y a s p ecif ic p erc en t c o m ple te . T his in fo rm atio n is in valu ab le to th e c o ntr o l p ro cess o f m ea su rin g s c h ed ule a n d c o st p erfo rm an ce— fo r e x am ple , p ay m en ts f o r c o ntr a ct w ork a re p aid o n “ p erc en t c o m ple te ” b asis . D efin in g a ta sk w ith c le ar d efin ab le s ta rt a n d e n d p oin ts a n d in te rm ed ia te p oin ts e n han ces th e c h an ces o f e arly d ete ctio n o f p ro ble m s, c o rre ctiv e a ctio n, a n d o n-tim e p ro je ct c o m ple tio n. T ypes o f C osts A ssu m in g w ork p ack ag es a re d efin ed , d eta ile d c o st e stim ate s c an b e m ad e. H ere a re t y pic al k in ds o f c o sts f o und i n a p ro je ct: 1 . D ir e ct c o sts a . L ab or b . M ate ria ls c . E quip m en t d . O th er 2 . D ir e ct p ro je ct o verh ead c o sts 3 . G en era l a n d a d m in is tr a tiv e ( G & A) o verh ead c o sts T he to ta l p ro je ct c o st e stim ate is b ro ken d ow n in th is fa sh io n to sh arp en th e c o ntr o l p ro cess a n d im pro ve d ecis io n m ak in g.

D ir e ct C osts 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 4/4 4 T hese c o sts a re c le arly c h arg eab le to a s p ecif ic w ork p ack ag e. D ir e ct c o sts c an b e in flu en ced b y th e p ro je ct m an ag er, pro je ct te am , an d in div id uals im ple m en tin g th e w ork pack ag e. T hese co sts re p re se n t re al cash o utf lo w s a n d m ust b e p aid a s t h e p ro je ct p ro gre sse s; t h ere fo re , d ir e ct c o sts a re u su ally s e p ara te d f ro m o verh ead c o sts . L ow er- le v el p ro je ct r o llu ps f re q uen tly i n clu de o nly d ir e ct c o sts . 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 5/4 4 P ag e 1 45 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . F IG URE 5 .5 C on tr a ct B id S um mary C osts D ir e ct P ro je ct O verh ea d C osts D ir e ct o verh ead r a te s m ore c lo se ly p in poin t w hic h r e so urc es o f th e o rg an iz atio n a re b ein g u se d in th e p ro je ct.

D ir e ct p ro je ct o verh ea d c o sts c a n b e t ie d t o p ro je ct d eliv era b le s o r w ork p ack ag es. E xam ple s i n clu de t h e s a la ry o f th e p ro je ct m an ag er a n d te m pora ry r e n ta l s p ace f o r th e p ro je ct te am . A lth ough o verh ead is n ot a n im med ia te o ut- o f-p ock et e x pen se , it is r e a l a n d m ust b e c o vere d in th e lo ng r u n if th e f ir m is to r e m ain v ia b le . T hese r a te s a re u su ally a ra tio o f th e d olla r v alu e o f th e re so urc es u se d — e.g ., d ir e ct la b or, m ate ria ls , eq uip m en t. F or e x am ple , a d ir e ct l a b o r b urd en r a te o f 2 0 p erc en t w ould a d d a d ir e ct o verh ead c h arg e o f 2 0 p erc en t t o t h e d ir e ct l a b or c o st e stim ate . A d ir e ct c h arg e r a te o f 5 0 p erc en t f o r m ate ria ls w ould c arry a n a d ditio nal 5 0 p erc en t c h arg e t o th e m ate ria l c o st e stim ate . S ele ctiv e d ir e ct o verh ead c h arg es p ro vid e a m ore a ccu ra te p ro je ct (jo b o r w ork p ack ag e) c o st, r a th er t h an u sin g a b la n ket o verh ead r a te f o r t h e w hole p ro je ct.

G en era l a n d A dm in is tr a tiv e ( G & A) O ve rh ea d C osts T hese r e p re se n t o rg an iz atio n c o sts th at a re n ot d ir e ctly lin ked to a s p ecif ic p ro je ct. T hese c o sts a re c arrie d f o r t h e d ura ti o n o f th e p ro je ct. E xam ple s in clu de o rg an iz atio n co sts acro ss all p ro ducts an d p ro je cts su ch as a d vertis in g, a cco untin g, a n d s e n io r m an ag em en t a b ove th e p ro je ct le v el. A llo ca tio n o f G & A c o sts v arie s f ro m o rg an iz atio n to o rg an iz atio n. H ow ev er, G & A c o sts a re u su ally a llo cate d a s a p erc en t o f to ta l d ir e ct c o st, o r a p erc en t o f t h e t o ta l o f a s p ecif ic d ir e ct c o st s u ch a s l a b or, m ate ria ls , o r e q uip m en t. G iv en t h e t o ta ls o f d ir e ct a n d o verh ead c o sts f o r i n div id ual w ork p ack ag es, i t i s p ossib le t o c u m ula te t h e c o sts f o r a n y d eliv era b le o r fo r th e e n tir e p ro je ct. A p erc en ta g e c an b e a d ded fo r p ro fit if y ou a re a c o ntr a cto r. A b re ak dow n o f c o sts f o r a p ro pose d c o ntr a ct b id i s p re se n te d i n F ig ure 5 .5 . P erc ep tio ns o f c o sts a n d b udgets v ary d ep en din g o n th eir u se rs . T he p ro je ct m an ag er m ust b e v ery a w are o f t h ese d if f e re n ces w hen se ttin g u p th e p ro je ct b udget a n d w hen c o m munic atin g th ese d if f e re n ces to o th ers .

F ig ure 5.6 dep ic ts th ese dif f e re n t perc ep tio ns. T he pro je ct m an ag er can co m mit co sts m onth s befo re th e r e so urc e i s u se d . T his i n fo rm atio n i s u se fu l t o t h e f in an cia l o ff ic er o f t h e o rg an iz atio n i n f o re castin g f u tu re c ash o utf lo w s. T he p ro je ct m an ag er is in te re ste d in w hen th e b udgete d c o st is e x pecte d to o ccu r, a n d w hen th e b udgete d c o st a ctu ally is c h arg ed ( e arn ed ); th e r e sp ectiv e tim in gs o f th ese tw o c o st f ig ure s a re u se d to m easu re p ro je ct s c h ed ule a n d c o st v aria n ces. 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 6/4 4 P ag e 1 46 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . F IG URE 5 .6 T hre e V ie w s o f C ost R efin in g E stim ate s A s desc rib ed earlie r in C hap te r 4 , deta ile d w ork pack ag e estim ate s are ag gre g ate d an d “ro lle d up” by d eliv era b le to e stim ate th e to ta l d ir e ct c o st o f th e p ro je ct. S im ila rly , e stim ate d d ura tio ns a re e n te re d in to th e p ro je ct n etw ork to e sta b lis h th e p ro je ct s c h ed ule a n d d ete rm in e th e o vera ll d ura tio n o f th e p ro je ct. E xperie n ce t e lls u s th at f o r m an y p ro je cts th e to ta l e stim ate s d o n ot m ate ria liz e a n d th e a c tu al c o sts a n d s c h ed ule o f s o m e p ro je cts s ig nif ic an tly e x ceed o rig in al w ork p ack ag e– base d e stim ate s. I n o rd er t o c o m pen sa te f o r t h e p ro ble m o f a ctu al c o st a n d s c h ed ule e x ceed in g e stim ate s, s o m e p ro je ct m an ag ers a d ju st to ta l c o sts b y s o m e m ultip lie r ( i.e ., t o ta l e stim ate d c o sts × 1 .2 0). T he p ra cti c e o f ad ju stin g o rig in al estim ate s b y 2 0 o r ev en 1 00 p erc en t b eg s th e q uestio n o f w hy, afte r i n vestin g s o m uch tim e a n d e n erg y o n d eta ile d e stim ate s, c o uld th e n um bers b e s o f a r o ff ? T here a re a n um ber o f re a so ns fo r th is , m ost o f w hic h c an b e tr a ced to th e e stim atin g p ro cess a n d th e in here n t u ncerta in ty o f p re d ic tin g t h e f u tu re . S om e o f t h ese r e aso ns a re d is c u sse d b elo w . I n te ra ctio n c o sts a re h id den i n e stim ate s. A cco rd in g t o t h e g uid elin es, e ach t a sk e stim ate i s s u ppose d t o b e d one i n dep en den tly . H ow ev er, t a sk s a re r a re ly c o m ple te d i n a v acu um . W ork o n o ne t a sk i s d ep en den t u pon p rio r t a sk s, a n d t h e h an d-o ff s b etw een t a sk s r e q uir e t im e a n d a tte n tio n. F or e x am ple , p eo ple w ork in g o n p ro to ty pe d ev elo pm en t n eed t o i n te ra ct w ith d esig n e n gin eers a fte r t h e d esig n i s c o m ple te d , w heth er t o s im ply a sk c la rif y in g q uestio ns o r t o m ak e a d ju stm en ts i n t h e o rig in al d esig n. S im ila rly , t h e t im e n ecessa ry t o c o ord in ate a ctiv itie s i s t y pic ally n ot r e fle cte d i n i n dep en den t e stim ate s. C oord in atio n i s r e fle cte d i n m eetin gs a n d b rie fin gs a s w ell a s t im e n ecessa ry t o r e so lv e d is c o nnects b etw een t a sk s. T im e, a n d t h ere fo re c o st, d ev ote d t o m an ag in g i n te ra ctio ns r is e s e x ponen tia lly a s t h e n um ber o f p eo ple a n d d if f e re n t d is c ip lin es i n volv ed i n cre ase s o n a p ro je ct.

N orm al c o n dit io n s d o n ot a p ply . E stim ate s a re s u ppose d t o b e b ase d o n n orm al c o nditio ns. W hile t h is i s a g ood s ta rtin g p oin t, i t r a re ly h old s t r u e i n r e al l if e . 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 7/4 4 P ag e 1 47 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . T his i s e sp ecia lly t r u e w hen i t c o m es t o t h e a v aila b ility o f r e so urc es. R eso urc e s h orta g es, w heth er i n t h e f o rm o f p eo ple , e q uip m en t, o r m ate ria ls , c an e x te n d o rig in al e stim ate s. F or e x am ple , u nder n orm al c o nditio ns f o ur b ulld ozers a re t y pic ally u se d t o c le ar a c erta in s ite s iz e i n f iv e d ay s, b ut t h e a v aila b ility o f o nly t h re e b ulld ozers w ould e x te n d t h e t a sk d ura tio n t o e ig ht d ay s.

S im ila rly , t h e d ecis io n t o o uts o urc e c erta in t a sk s c an i n cre ase c o sts a s w ell a s e x te n d t a sk d ura tio ns s in ce t im e i s a d ded t o a cclim atin g o uts id ers t o t h e p artic u la rs o f t h e p ro je ct a n d t h e c u ltu re o f t h e o rg an iz atio n.

T hin gs g o w ro n g o n p ro je cts . D esig n f la w s a re r e v eale d a fte r t h e f a ct, e x tr e m e w eath er c o nditio ns o ccu r, a ccid en ts h ap pen , a n d s o f o rth . A lth ough y ou s h ould n't p la n f o r t h ese r is k s t o h ap pen w hen e stim atin g a p artic u la r t a sk , t h e l ik elih ood a n d i m pact o f s u ch e v en ts n eed t o b e c o nsid ere d .

C han ges i n p ro je ct s c o p e a n d p la n s. A s o ne g ets f u rth er a n d f u rth er i n to t h e p ro je ct, a m an ag er o bta in s a b ette r u nders ta n din g o f w hat n eed s t o b e d one t o a cco m plis h t h e p ro je ct. T his m ay l e ad t o m ajo r c h an ges i n p ro je ct p la n s a n d c o sts . L ik ew is e , i f t h e p ro je ct i s a c o m merc ia l p ro je ct, c h an ges o fte n h av e t o b e m ad e m id str e am t o r e sp ond t o n ew d em an ds b y t h e c u sto m er a n d/o r c o m petitio n. U nsta b le p ro je ct s c o pes a re a m ajo r s o urc e o f c o st o verru ns. W hile e v ery e ff o rt s h ould b e m ad e u p f ro nt t o n ail d ow n t h e p ro je ct s c o pe, i t i s b eco m in g i n cre asin gly d if f ic u lt t o d o s o i n o ur r a p id ly c h an gin g w orld .

O verly o p tim is tic . T here i s s o lid r e se arc h i n dic atin g t h at t h ere i s a t e n den cy i n p eo ple t o o vere stim ate h ow q uic k ly t h ey c an g et t h in gs d one a n d u ndere stim ate h ow l o ng i t w ill t a k e t h em t o c o m ple te t a sk s.

T his w ould b e e sp ecia lly t r u e i n o rg an iz atio nal c u ltu re s w here p ad din g i s n ot t h e n orm . S ee S nap sh ot f ro m P ra ctic e: M eg a P ro je cts : A S pecia l C ase .

S tr a te g ic m is r e p re se n ta tio n . T here i s g ro w in g e v id en ce t h at s o m e p ro je ct p ro m ote rs u ndere stim ate t h e c o sts o f p ro je cts a n d o vere stim ate p ro je ct b en efits i n o rd er t o w in a p pro val. T his a p pears t o b e p artic u la rly t r u e f o r l a rg e-s c ale p ublic w ork s p ro je cts w hic h h av e a n oto rio us h ab it o f c o m in g i n w ay o ver b udget ( re m em ber t h e e arlie r C ouncil F um es S nap sh ot) . T he r e ality i s t h at f o r m an y p ro je cts n ot a ll o f t h e i n fo rm atio n n eed ed t o m ak e a c cu ra te e stim ate s i s a v aila b le , a n d it is im possib le to p re d ic t th e fu tu re . T he ch alle n ge is fu rth er co m pound ed b y h um an n atu re an d th e p olitic al d ynam ic s a sso cia te d w ith g ain in g p ro je ct a p pro val. T he d ile m ma is th at w ith out s o lid e stim ate s, th e c re d ib ility of th e pro je ct pla n is ero ded . D ead lin es beco m e m ean in gle ss, budgets beco m e ru bbery , an d a cco unta b ility b eco m es p ro ble m atic . C halle n ges s im ila r to th ose d esc rib ed a b ove w ill in flu en ce th e fin al tim e a n d c o st e stim ate s. E ven w ith th e b est estim atin g eff o rts , it m ay be necessa ry to re v is e estim ate s base d on re le v an t in fo rm atio n p rio r to e sta b lis h in g a b ase lin e s c h ed ule a n d b udget. E ff e ctiv e o rg an iz atio ns a d ju st e stim ate s o f s p ecif ic t a sk s o nce r is k s, r e so urc es, a n d p artic u la rs o f t h e s itu atio n h av e b een m ore c le arly d efin ed . T hey r e co gniz e th at th e r o lle d u p e stim ate s g en era te d f ro m a d eta ile d e stim ate b ase d o n t h e W BS a re j u st t h e s ta rtin g p oin t. A s t h ey d elv e f u rth er i n to t h e p ro je ct- p la n nin g p ro cess, t h ey m ak e a p pro pria te re v is io ns b oth in th e tim e an d co st o f sp ecif ic activ itie s. T hey fa cto r th e fin al assig nm en t o f r e so urc es in to th e p ro je ct b udget a n d s c h ed ule . F or e x am ple , w hen th ey r e aliz e th at o nly th re e in ste ad o f f o ur b ulld ozers a re a v aila b le to c le ar a s ite , th ey a d ju st b oth th e tim e a n d c o st o f th at a ctiv ity . T hey a d ju st e stim ate s t o a cco unt f o r s p ecif ic a ctio ns to m itig ate p ote n tia l r is k s o n th e p ro je ct. F or e x am ple , to r e d uce th e c h an ces o f d esig n co de erro rs , th ey w ould ad d th e co st of in dep en den t te ste rs to th e sc h ed ule an d budget. F in ally , o rg an iz atio ns a d ju st e stim ate s to ta k e in to a cco unt a b norm al c o nditio ns. F or e x am ple , if s o il s a m ple s re v eal e x cessiv e g ro und w ate r, t h en t h ey a d ju st f o undatio n c o sts a n d t im es. 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 8/4 4 P ag e 1 48 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . R ese arc h H ig hlig ht M eg a P ro je cts : A S pecia l C ase T he re co rd o f c o st e stim ate s fo r m eg a p ro je c ts is n ot s te lla r. P opula r e x am ple s o f la rg e c o st o verru ns in clu de th e S ydney O pera H ouse , th e C hunnel c o nnectin g E ngla n d a n d F ra n ce, a n d B osto n's a rte ry /tu nnel “ B ig D ig .” I n a s tu dy o f g overn m en t- a p pro ved in fra str u ctu re p ro je cts , B en t F ly vbje rg f o und c o sts f o r b rid ges a n d tu nnels , f o r r o ad , a n d f o r r a il t o b e u ndere stim ate d 3 3.8 % , 2 0.4 % , a n d 4 4.7 % , r e sp ectiv ely , f ro m i n itia l b ase lin e e stim ate s t o r e ality . 1 F ly vbje rg s u ggeste d “ … g overn m en t o ff ic ia ls s h ould n ot t r u st c o st e stim ate s p re se n te d b y p ro je ct p ro m ote rs w hose m ain a m bitio n is to b uild a p ro je ct fo r th eir p riv ate o r p o litic al g ain . T hese e stim ate s te n d to b e o verly o ptim is tic .

A dvocate s k now th at h ig h c o sts re d uce th e li k elih ood o f fu ndin g.” T he is su e c o m es d ow n to h ow to m itig ate th ese i n flu en ces i n o ur e stim ate s. O ne a p pro ach lie s in u sin g a n “ ex te rn al v ie w ” b ase d o n a ctu al o utc o m es o f s im il a r p ro je cts c o m ple te d in th e p ast. I t i s c alle d R efe re n ce C la ss F ore ca stin g ( R CF) a n d u se s F ly vbje rg 's r e se arc h a p pro ach . R EFE R EN CE C LA SS F O REC AST IN G M ETH ODO LO GY F undam en ta lly , h is m eth od i s s im ple a n d s tr a ig htf o rw ard a n d i n volv es t h re e m ajo r s te p s: 1 . S ele ct a r e fe re n ce c la ss o f p ro je cts s im ila r t o y o ur p ote n tia l p ro je ct, f o r e x am ple , c arg o s h ip s o r b rid ges. 2 . C olle ct a nd a rra nge o utc o m e d ata a s a d is tr ib utio n. C re ate a d is tr ib utio n o f c o st o verru ns a s a p erc en t o f o rig in al p ro je ct e stim ate ( lo w t o h ig h). 3 . U se t h e d is tr ib utio n d ata t o a rriv e a t a r e a lis ti c f o re ca st. C om pare t h e o rig in al c o st e stim ate f o r t h e p ro je ct w ith t h e r e fe re n ce c la ss p ro je cts . ( F or e x am ple , a sk a d vocate w hat s tr o ng e v id en ce d o y ou h av e t h at y our p ro je ct w ill n ot f o llo w t h e p ro je cts i n t h e r e fe re n ce c la ss? ) A PPL IC AT IO N F ly vbje rg 's s e m in al s tu die s s e rv ed a s a n im petu s fo r th e u se o f R efe re n ce C la ss F ore castin g. F or e x am ple , th e U K D ep artm en t f o r T ra n sp ort a n d H M T re asu ry e m plo ys th e m eth od a s p art o f p ro je ct a p pra is a l f o r la rg e tr a n sp orta tio n p ro je cts u n der th eir ju ris d ic tio n. T he A m eric an P la n nin g A sso cia tio n (A PA ) re co m men ds th e R CF m eth od an d s tr o ngly s u ggests p la n ners s h ould n ev er re ly s o le ly o n c o nven tio nal fo re castin g te ch niq ues w hen m ak in g fo re casts .

T he A ACE I n te rn atio nal ( th e A sso cia tio n f o r th e A dvan cem en t o f C ost E ngin eerin g) r e co m men ds th e u se o f R CF a s a v alid atio n o f c o st e stim ate s. R efe re n ce c la ss fo re castin g c an e asily b e a p plie d to p ro je cts su ch a s sp ort e v en ts , r e se arc h an d d ev elo p m en t, in fo rm atio n te ch nolo gy sy ste m s, p ow er p la n ts an d d am s, co nstr u ctio n, sp ort are n as, m ovie s, e n erg y e x plo ra tio n, s p ace e x plo ra tio n, a n d m an y m ore c la sse s. 2 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 2 9/4 4 © M ic h ael D wyer/A la m y B EN EFIT S T he b en efits o f r e fe re n ce c la ss f o re castin g a re c o m pellin g: O uts id e e m pir ic al d ata m itig ate s h um an b ia s.

P olitic al, s tr a te g ic , a n d p ro m ote r f o rc es h av e d if f ic u lty i g norin g o uts id e r e fe re n ce c la ss f o re cast i n fo rm atio n.

S erv es a s r e ality c h eck f o r f u ndin g l a rg e p ro je cts .

H elp s e x ecu tiv es a v oid u nso und o ptim is m .

L ead s t o i m pro ved a cco unta b ility .

P ro vid es b asis f o r p ro je ct c o ntin gen cy f u nds. F U TU RE A PPL IC AT IO N B EY O ND I N FR AST R UCTU RE G iv en c o st o verru ns i n n in e o ut o f t e n m eg a p ro je cts , i n accu ra cie s h av e t o b e r e d uced t o i m pro ve c h an ces o f f u ndin g.

T he u se o f R efe re n ce C la ss F ore castin g i s i n cre asin g a s g o vern m en ts a n d o rg an iz atio ns r e q uir e t h e m eth od b e u se d t o t e m per p ro je ct a d vocate e stim ate s a n d r e d uce i n accu ra cie s.

1 S ee f o r e x am ple : L ovallo , D ., a n d D . K ah nem an , “ D elu sio ns o f S uccess: H ow O ptim is m U nderm in es E xecu tiv es’ D ecis io ns,” H arv a rd B usin ess R evie w , J u ly 2 003, p p. 5 6–63; B ueh le r, R ., D . G rif f e n , a n d M . R oss, “ E xplo rin g t h e ‘ P la n nin g F all a cy ’: W hy P eo ple U ndere stim ate T heir T ask C om ple tio n T im es,” J o urn al o f P erso nality a nd S ocia l P sy ch olo gy, 6 7 ( 3 ), p p. 3 66–381. 2 F ly vbje rg , B en t, “ F ro m N obel P riz e t o P ro je ct M an ag em en t: G ett in g R is k s R ig ht,” P ro je ct M anagem en t J o urn al, A ugust 2 006, p p. 5 –15. 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 0/4 4 P ag e 1 49 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . T here w ill a lw ay s b e so m e m is ta k es, o m is sio ns, a n d a d ju stm en ts th at w ill re q uir e a d ditio nal c h an ges in e stim ate s. F ortu nate ly e v ery p ro je ct sh ould h av e a c h an ge m an ag em en t sy ste m in p la ce to acco m modate th ese situ atio ns an d an y im pact on th e pro je ct base lin e. C han ge m an ag em en t a n d c o ntin gen cy f u nds w ill b e d is c u sse d l a te r i n C hap te r 7 . C re atin g a D ata b ase f o r E stim atin g T he b est w ay to im pro ve e stim ate s is to c o lle ct a n d a rc h iv e d ata o n p ast p ro je ct e stim ate s a n d a ctu als . S av in g h is to ric al data — estim ate s an d actu als — pro vid es a know le d ge base fo r im pro vin g pro je ct tim e an d co st e stim atin g. Cre atin g an estim atin g data b ase is a “b est pra ctic e” am ong le ad in g pro je ct m an ag em en t o rg an iz atio ns. S om e o rg an iz atio ns h av e la rg e e stim atin g d ep artm en ts o f p ro fe ssio nal e stim ato rs — e.g ., B oein g, I B M — th at h av e d ev elo ped la rg e t im e a n d co st d ata b ase s. O th ers co lle ct th ese d ata th ro ugh th e p ro je ct o ff ic e. T his d ata b ase a p pro ach a llo w s th e p ro je ct e stim ato r to s e le ct a s p ecif ic w ork p ack ag e ite m fro m th e d ata b ase fo r i n clu sio n. T he e stim ato r th en m ak es a n y n ecessa ry a d ju stm en ts c o ncern in g th e m ate ria ls , la b or, a n d e q uip m en t.

O f c o u rs e a n y ite m s n ot fo und in th e d ata b ase c an b e a d ded to th e p ro je ct— an d u ltim ate ly to th e d ata b ase if d esir e d . A gain , th e q u ality o f th e d ata b ase e stim ate s d ep en ds o n th e e x perie n ce o f th e e stim ato rs , b ut o ver tim e t h e d ata q uality s h ou ld im pro ve. S uch s tr u ctu re d d ata b ase s s e rv e a s f e ed back f o r e stim ato rs a n d a s b en ch m ark s f o r c o st a n d tim e fo r e ach p ro je ct. In a d ditio n, c o m paris o n o f e stim ate a n d a ctu al fo r d if f e re n t p ro je cts c an s u ggest th e d eg re e o f r is k in here n t in e stim ate s. S ee F ig ure 5 .7 f o r th e s tr u ctu re o f a d ata b ase s im ila r to th ose f o und i n p ra ctic e.

F IG URE 5 .7 E stim atin g D ata b ase T em pla te s 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 1/4 4 P ag e 1 50 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . S um mary Q uality tim e a n d c o st e stim ate s a re th e b ed ro ck o f p ro je ct c o ntr o l. P ast e x perie n ce is th e b est s ta rtin g p oin t f o r t h ese estim ate s. T he quality of estim ate s is in flu en ced by oth er fa cto rs su ch as peo ple , te ch nolo gy, an d d ow ntim es. T he key fo r gettin g estim ate s th at re p re se n t re alis tic av era g e tim es an d co sts is to hav e an o rg an iz atio n c u ltu re th at a llo w s e rro rs in e stim ate s w ith out in crim in atio ns. I f tim es r e p re se n t a v era g e tim e, w e s h ould e x pect th at 5 0 p erc en t w ill b e le ss th an th e e stim ate a n d 5 0 p erc en t w ill e x ceed th e e stim ate . T he u se o f t e am s th at a re h ig hly m otiv ate d c an h elp in k eep in g ta sk tim es a n d c o sts n ear th e a v era g e. F or th is r e aso n, it is c ru cia l t o g et t h e t e am t o b uy i n to t im e a n d c o st e stim ate s. U sin g to p-d ow n e stim ate s is g ood fo r in itia l a n d s tr a te g ic d ecis io n m ak in g o r in s itu atio ns w here th e c o sts a sso cia te d w ith d ev elo pin g b ette r e stim ate s h av e little b en efit. H ow ev er, in m ost c ase s th e b otto m -u p a p pro ach t o e stim atin g is p re fe rre d a n d m ore re lia b le b ecau se it a sse sse s e ach w ork p ack ag e, ra th er th an th e w hole p ro je ct, se ctio n, or deliv era b le of a pro je ct. E stim atin g tim e an d co sts fo r each w ork pack ag e fa cilita te s d ev elo pm en t o f th e p ro je ct s c h ed ule a n d a tim e-p hase d b udget, w hic h a re n eed ed to c o ntr o l th e p ro je ct a s it is i m ple m en te d . U sin g th e estim atin g g u id elin es w ill h elp elim in ate m an y co m mon m is ta k es m ad e b y th ose u nacq uain te d w ith estim atin g tim es an d co sts fo r p ro je ct co ntr o l. E sta b lis h in g a tim e an d co st estim atin g d ata b ase f its w ell w ith t h e l e arn in g o rg an iz atio n p hilo so phy. T he le v el o f tim e a n d c o st d eta il s h ould f o llo w th e o ld s a y in g o f “ n o m ore th an is n ecessa ry a n d s u ff ic ie n t.” M an ag ers m ust re m em ber to d if f e re n tia te b etw een c o m mitte d o utla y s, a ctu al c o sts , a n d s c h ed ule d c o sts . It is w ell k now n th at u pfro nt e ff o rts in c le arly d efin in g p ro je ct o bje ctiv es, s c o pe, a n d s p ecif ic atio ns v astly im pro ve t im e a n d c o st e stim ate a ccu ra cy. F in ally , h ow estim ate s are g ath ere d an d h ow th ey are u se d can aff e ct th eir u se fu ln ess fo r p la n nin g an d c o ntr o l. The te am clim ate , org an iz atio n cu ltu re , an d org an iz atio n str u ctu re can str o ngly in flu en ce th e i m porta n ce a tta ch ed t o t im e a n d c o st e stim ate s a n d h ow t h ey a re u se d i n m an ag in g p ro je cts . K ey T erm s A pportio nm en t, 1 37 B otto m -u p e stim ate s, 1 34 D elp hi M eth od, 1 36 D ir e ct c o sts , 1 44 F unctio n p oin ts , 1 37 L earn in g c u rv es, 1 39 O verh ead c o sts , 1 45 P ad din g e stim ate s, 1 30 P hase e stim atin g, 1 41 R an ge e stim atin g, 1 40 R atio m eth ods, 1 36 R efe re n ce C la ss F ore castin g, 1 48 T em pla te m eth od, 1 39 T im e a n d c o st d ata b ase s, 1 45 T op-d ow n e stim ate s, 1 34 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 2/4 4 R ev ie w Q uestio n s 1 . W hy a re a ccu ra te e stim ate s c ritic al t o e ff e ctiv e p ro je ct m an ag em en t? 2 . H ow d oes t h e c u ltu re o f a n o rg an iz atio n i n flu en ce t h e q uality o f e stim ate s? 3 . W hat a re t h e d if f e re n ces b etw een b otto m -u p a n d t o p-d ow n e stim atin g a p pro ach es? U nder w hat c o nditio ns w ould y ou p re fe r o ne o ver t h e o th er? 4 . W hat a re t h e m ajo r t y pes o f c o sts ? W hic h c o sts a re c o ntr o lla b le b y t h e p ro je ct m an ag er? E xerc is e s 1 . C alc u la te t h e d ir e ct c o st o f l a b or f o r a p ro je ct t e am m em ber u sin g t h e f o llo w in g d ata : H ourly r a te : $ 40/h r H ours n eed ed : 8 0 O verh ead r a te : 4 0% 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 3/4 4 P ag e 1 51 2 . P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . M rs . T ols to y a n d h er h usb an d, S erg e, a re p la n nin g t h eir d re am h ouse . T he l o t f o r t h e h ouse s its h ig h o n a h ill w ith a b eau tif u l v ie w o f t h e A ppala ch ia n M ounta in s.

T he p la n s s h ow t h e s iz e o f t h e h ouse t o b e 2 ,9 00 s q uare f e et. T he a v era g e p ric e f o r a l o t a n d h ouse s im ila r t o t h is o ne h as b een $ 120 p er s q uare f o ot. F ortu nate ly , S erg e i s a r e tir e d p lu m ber a n d f e els h e c an s a v e m oney b y i n sta llin g t h e p lu m bin g h im se lf . M rs . T ols to y f e els s h e c an t a k e c are o f t h e i n te rio r d eco ra tin g. T he f o llo w in g a v era g e c o st i n fo rm atio n i s a v aila b le f ro m a l o cal b an k t h at m ak es l o an s t o l o cal c o ntr a cto rs a n d d is p ers e s p ro gre ss p ay m en ts to c o ntr a cto rs w hen sp ecif ic ta sk s a re v erif ie d a s c o m ple te . 2 4% E xcav atio n a n d f ra m in g c o m ple te 8 % R oof a n d f ir e p la ce c o m ple te 3 % W ir in g r o ughed i n 6 % P lu m bin g r o ughed i n 5 % S id in g o n 1 7% W in dow s, i n su la tio n, w alk s, p la ste r, a n d g ara g e c o m ple te 9 % F urn ace i n sta lle d 4 % P lu m bin g f ix tu re s i n sta lle d 1 0% E xte rio r p ain t, l ig ht f ix tu re s i n sta lle d , f in is h h ard w are i n sta lle d 6 % C arp et a n d t r im i n sta lle d 4 % I n te rio r d eco ra tin g 4 % F lo ors l a id a n d f in is h ed a . W hat i s t h e e stim ate d c o st f o r t h e T ols to y's h ouse i f t h ey u se c o ntr a cto rs t o c o m ple te a ll o f t h e h ouse ? b . E stim ate w hat t h e c o st o f t h e h ouse w ould b e i f t h e T ols to ys u se t h eir t a le n ts t o d o s o m e o f t h e w ork t h em se lv es. 3 . B elo w i s a p ro je ct W BS w ith c o st a p portio ned b y p erc en ts . I f t h e t o ta l p ro je ct c o st i s e stim ate d t o b e $ 600,0 00, w hat a re t h e e stim ate d c o sts f o r t h e f o llo w in g d eliv era b le s? a . D esig n? b . P ro gra m min g? c . I n -h ouse t e stin g?

W hat w eak nesse s a re i n here n t i n t h is e stim atin g a p pro ach ? 4 . F ir e w all P ro je ct X T. U sin g t h e “ co m ple x ity w eig htin g” s c h em e s h ow n i n E xerc is e 5 .3 a n d t h e f u nctio n p oin t c o m ple x ity w eig hte d t a b le s h ow n b elo w , e stim ate t h e t o ta l f u nctio n p oin t c o unt.

A ssu m e h is to ric al d ata s u ggest f iv e f u nctio n p oin ts e q ual o ne p ers o n a m onth a n d s ix p eo ple c an w ork o n t h e p ro je ct. a . W hat i s t h e e stim ate d p ro je ct d ura tio n? b . I f 2 0 p eo ple a re a v aila b le f o r t h e p ro je ct, w hat i s t h e e stim ate d p ro je ct d ura tio n? c . I f t h e p ro je ct m ust b e c o m ple te d i n s ix m onth s, h ow m an y p eo ple w ill b e n eed ed f o r t h e p ro je ct? 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 4/4 4 P ag e 1 52 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . E X ER CIS E 5 .3 W BS F ig u re R efe re n ces B ueh le r, R ., D . G rif f e n , a n d M . R oss, “ E xplo rin g t h e ‘ P la n nin g F alla cy ’: W hy P eo ple U ndere stim ate T heir T ask C om ple tio n T im es,” J o urn al o f P erso nality a nd S ocia l P sy ch olo gy, v ol. 6 7 ( 3 ), p p. 3 66–381. D alk ey, N . C ., D . L . R ourk e, R . L ew is , a n d D . S nyder, S tu die s i n t h e Q uality o f L ife : D elp hi a nd D ecis io n M akin g ( L ex in gto n, M A: L ex in gto n B ooks, 1 972). F ly vbje rg , B en t, “ F ro m N obel P riz e t o P ro je ct M an ag em en t: G ettin g R is k s R ig ht,” P ro je ct M anagem en t J o urn al, A ugust 2 006, p p. 5 –15; F ly vbje rg , B en t, “ C urb in g O ptim is m B ia s a n d S tr a te g ic M is re p re se n ta tio n i n P la n nin g: R efe re n ce C la ss F ore castin g i n P ra ctic e,” E uro pea n P la nnin g S tu die s, v ol. 1 6, N o. 1 ( J a n uary 2 008), p p. 3 –21; a n d F ly vbje rg , B en t, N . B ru zeliu s, a n d W . R oth en gatte r, M eg a P ro je cts a nd R is k : A n A nato m y o f A m bitio n ( C am brid ge P re ss, 2 003). G ra y, N . S ., “ S ecre ts t o C re atin g t h e E lu siv e ‘ A ccu ra te E stim ate ,’ ” P M N etw ork , 1 5 ( 8 ) A ugust 2 001, p . 5 6.

J e ff e ry , R ., G . C . L ow , a n d M . B arn es, “ A C om paris o n o f F unctio n P oin t C ountin g T ech niq ues,” I E EE T ra nsa ctio ns o n S oftw are E ngin eerin g, 1 9 ( 5 ) 1 993, p p. 5 29–32. J o nes, C ., A pplie d S oftw are M ea su re m en t ( N ew Y ork : M cG ra w -H ill, 1 991). J o nes, C ., E stim atin g S oftw are C osts ( N ew Y ork : M cG ra w -H ill, 1 998). K harb an da, O . P ., a n d J . K . P in to , W hat M ade G ertie G allo p: L ea rn in g f r o m P ro je ct F ailu re s ( N ew Y ork : V on N ostr a n d R ein hold , 1 996). 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 5/4 4 L ovallo , D ., a n d D . K ah nem an , “ D elu sio ns o f S uccess: H ow O ptim is m U nderm in es E xecu tiv es’ D ecis io ns,” H arv a rd B usin ess R evie w , J u ly 2 003, p p. 5 6–63. M ag ne, E ., K . E m hje lle n m , a n d P . O sm undse n , “ C ost E stim atio n O verru ns i n t h e N orth S ea,” P ro je ct M anagem en t J o urn al 3 4 ( 1 ) 2 003, p p. 2 3–29. 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 6/4 4 P ag e 1 53 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . M cL eo d, G ., a n d D . S m ith , M anagin g I n fo rm atio n T ech nolo gy P ro je cts ( C am brid ge, M A: C ours e T ech nolo gy, 1 996).

M ilo se v ic , D . Z ., P ro je ct M anagem en t T o olB ox ( U pper S ad dle R iv er, N J: J o hn W ile y, 2 003), p . 2 29. P re ssm an , R . S ., S oftw are E ngin eerin g: A P ra ctitio ner's A ppro ach , 4 th e d . ( N ew Y ork : M cG ra w -H ill, 1 997).

S ym ons, C . R ., “ F unctio n P oin t A naly sis : D if f ic u ltie s a n d I m pro vem en ts ,” I E EE T ra nsa ctio ns o n S oftw are E ngin eerin g, 1 4 ( 1 ) 1 988, p p. 2 –11 . C ase S harp P rin tin g, A G T hre e y ears a g o th e S harp P rin tin g ( S P) s tr a te g ic m an ag em en t g ro up s e t a g oal o f h av in g a c o lo r la se r p rin te r a v aila b le fo r th e co n su m er an d sm all b usin ess m ark et fo r le ss th an $ 200. A fe w m onth s la te r th e se n io r m an ag em en t m et o ff -s ite t o d is c u ss t h e n ew p ro duct. T he r e su lts o f t h is m eetin g w ere a s e t o f g en era l t e ch nic al s p ecif ic atio ns alo ng w ith m ajo r deliv era b le s, a pro duct la u nch date , an d a co st estim ate base d on prio r e x perie n ce. S hortly afte rw ard , a m eetin g w as arra n ged fo r m id dle m an ag em en t ex pla in in g th e p ro je ct g oals , m ajo r r e sp onsib ilitie s, th e p ro je ct sta rt d ate , an d im porta n ce o f m eetin g th e p ro duct la u nch d ate w ith in th e co st e stim ate . M em bers o f a ll d ep artm en ts i n v olv ed a tte n ded t h e m eetin g. E xcite m en t w as h ig h. A lth ough e v ery one s a w th e ris k s a s h ig h , th e p ro m is e d re w ard s fo r th e c o m pan y a n d th e p ers o nnel w ere e m bla zo ned in th eir m in ds. A fe w p artic ip an ts q uestio ned th e le g itim acy o f th e p ro je ct d ura tio n a n d c o st e stim ate s. A c o uple o f R & D p eo ple w ere w orrie d a b out th e te ch nolo gy re q uir e d to p ro duce th e h ig h-q uality p ro duct fo r le ss th an $ 200. B ut g iv en th e e x cite m en t o f th e m om en t, e v ery one a g re ed th e p ro je ct w as w orth d oin g a n d d oab le . T he c o lo r l a se r p rin te r p ro je ct w as t o h av e t h e h ig hest p ro je ct p rio rity i n t h e c o m pan y. L au re n w as se le cte d to b e th e p ro je ct m an ag er. S he h ad 1 5 y ears o f ex perie n ce in p rin te r d esig n an d m an ufa ctu re , w hic h in clu ded su ccessfu l m an ag em en t o f se v era l p ro je cts re la te d to p rin te rs fo r c o m merc ia l m ark ets . S in ce s h e w as o ne o f th ose u nco m fo rta b le w ith th e p ro je ct c o st a n d tim e e stim ate s, s h e fe lt g ettin g g ood b otto m -u p tim e a n d c o st e stim ate s f o r th e d eliv era b le s w as h er f ir s t c o ncern . S he q uic k ly h ad a m eetin g w ith th e sig nif ic an t sta k eh old ers to cre ate a W BS id en tif y in g th e w ork p ack ag es an d o rg an iz atio nal u nit r e sp onsib le fo r im ple m en tin g th e w ork p ack ag es. L au re n str e sse d sh e w an te d tim e a n d c o st e stim ate s fro m t h ose w ho w ould d o th e w ork o r w ere th e m ost k now le d geab le , if p ossib le . G ettin g e stim ate s f ro m m ore th an o ne s o urc e w as e n co ura g ed . E stim ate s w ere d ue i n t w o w eek s. T he c o m pile d e stim ate s w ere p la ced in th e W BS/O BS. T he c o rre sp ondin g c o st e stim ate se em ed to b e in e rro r. T he co st estim ate w as $ 1,2 50,0 0 0 o ver th e se n io r m an ag em en t estim ate ; th is re p re se n ts ab out a 2 0 p erc en t o verru n! T he tim e e stim ate fro m th e d ev elo ped p ro je ct n etw ork w as o nly fo ur m onth s o ver th e to p m an ag em en t t im e e stim ate . A noth er m eetin g w as s c h ed ule d w ith t h e s ig nif ic an t 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 7/4 4 P ag e 1 54 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . s ta k eh old ers to c h eck th e e stim ate s a n d to b ra in sto rm fo r a lte rn ativ e so lu tio ns; th e c o st a n d t im e e stim ate s a p peare d to b e r e aso nab le . S om e o f th e s u ggestio ns f o r th e b ra in sto rm in g s e ssio n a re l is te d b elo w . C han ge s c o pe.

O uts o urc e t e ch nolo gy d esig n.

U se t h e p rio rity m atr ix ( fo und i n C hap te r 4 ) t o g et t o p m an ag em en t t o c la rif y t h eir p rio ritie s. P artn er w ith a n oth er o rg an iz atio n o r b uild a r e se arc h c o nso rtiu m t o s h are c o sts a n d t o s h are t h e n ew ly d ev elo ped t e ch nolo gy a n d p ro ductio n m eth ods.

C an cel t h e p ro je ct.

C om mis sio n a b re ak -e v en s tu dy f o r t h e l a se r p rin te r. V ery little in th e w ay o f c o ncre te sa v in gs w as id en tif ie d , a lth ough th ere w as c o nse n su s th at tim e c o uld b e c o m pre sse d t o t h e m ark et l a u nch d ate , b ut a t a d ditio nal c o sts . L au re n m et w ith th e m ark etin g ( C onnor) , p ro ductio n ( K im ), a n d d esig n ( G ag e) m an ag ers w ho y ie ld ed s o m e i d eas f o r c u ttin g c o sts , b ut n oth in g s ig nif ic an t e n ough t o h av e a l a rg e i m pact. G ag e r e m ark ed , “ I w ould n't w an t t o b e th e o ne to d eliv er th e m essa g e to to p m an ag em en t th at th eir c o st e stim ate is $ 1,2 50,0 00 o ff ! G ood lu ck , L au re n .” 1 . A t t h is p oin t, w hat w ould y ou d o i f y ou w ere t h e p ro je ct m an ag er? 2 . W as t o p m an ag em en t a ctin g c o rre ctly i n d ev elo pin g a n e stim ate ? 3 . W hat e stim atin g t e ch niq ues s h ould b e u se d f o r a m is sio n c ritic al p ro je ct s u ch a s t h is ? C ase P ost G ra d uatio n A dven tu re J o sh an d M ik e m et each oth er as ro om mate s durin g fre sh m en year at M acA lis te r C olle g e in S t. P au l, M in neso ta . D esp ite a ro ck y sta rt th ey b ecam e b est frie n ds. T hey are p la n nin g o n g oin g o n a tw o w eek a d ven tu re to geth er to c ele b ra te th eir g ra d uatio n in Ju ne. Jo sh h as n ev er b een to E uro pe a n d w an ts to v is it F ra n ce o r S pain . M ik e sp en t a se m este r a b ro ad in A arh us, D en m ark , a n d tr a v ele d e x te n siv ely in N orth ern E uro pe. E ven t h ough h e n ev er w en t t o F ra n ce o r S pain , M ik e w an ts t o g o t o s o m e p la ce m ore e x otic l ik e S outh A fric a o r V ie tn am . F or th e p ast w eek th ey h av e b een a rg uin g b ack a n d f o rth o ver w here th ey s h ould g o. J o sh a rg ues th at it w ill c o st to o m uch to f ly to S outh A fric a o r V ie tn am , w hile M ik e c o unte rs th at it w ill b e m uch c h eap er to tr a v el in V ie tn am o r S outh A fric a o nce th ey a re th ere . E ach o f th em a g re ed th at th ey c an s p en d n o m ore t h an $ 3,5 00 e ach o n t h e t r ip a n d c o uld b e g one f o r o nly t w o w eek s. O ne e v en in g w hen t h ey w ere a rg uin g w ith e ach o th er o ver b eers w ith f rie n ds, S ara s a id , “ W hy d on't y ou u se w hat y ou l e arn ed i n y our p ro je ct m an ag em en t c la ss t o d ecid e w hat t o d o?” J o sh a n d M ik e l o oked a t e ach o th er a n d a g re ed t h at m ad e p erfe ct s e n se . 1 . A ssu m e y ou a re e ith er M ik e o r J o sh ; h ow w ould y ou g o a b out m ak in g a d ecis io n u sin g p ro je ct m an ag em en t m eth odolo gy? 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 8/4 4 P ag e 1 55 2 . P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . L ookin g f ir s t a t o nly c o st, w hat d ecis io n w ould y ou m ak e? 3 . A fte r c o st, w hat o th er f a cto rs s h ould b e c o nsid ere d b efo re m ak in g a d ecis io n? A ppen dix 5 .1 L earn in g C urv es f o r E stim atin g A f o re cast e stim ate o f th e tim e r e q uir e d to p erfo rm a w ork p ack ag e o r ta sk is a b asic n ecessity f o r s c h ed ulin g t h e p ro je ct. In s o m e c ase s, th e m an ag er s im ply u se s ju dgm en t a n d p ast e x perie n ce to e stim ate w ork p ack ag e t im e, o r m ay u se h is to ric al r e co rd s o f s im ila r t a sk s. M ost m an ag ers a n d w ork ers in tu itiv ely k now th at im pro vem en t in th e a m ount o f tim e r e q uir e d to p erfo rm a t a sk o r g ro up o f ta sk s o ccu rs w ith r e p etitio n. A w ork er c an p erfo rm a ta sk b ette r/q uic k er th e s e co nd tim e a n d e ach s u ccee d in g tim e s h e/h e p erfo rm s it ( w ith out a n y te ch nolo gic al c h an ge). I t is th is p atte rn o f im pro vem en t t h at i s i m porta n t t o t h e p ro je ct m an ag er a n d p ro je ct s c h ed ule r. T his im pro v em en t f ro m r e p etitio n g en era lly r e su lts in a r e d uctio n o f la b or h ours f o r th e a cco m plis h m en t o f t a sk s an d re su lts in lo w er p ro je ct co sts . F ro m em pir ic al ev id en ce acro ss a ll in du str ie s, th e p att e rn o f th is i m pro vem en t h as b een q uan tif ie d in th e l e a rn in g c u rv e ( a ls o k now n a s im pro vem en t c u rv e, e x perie n ce c u rv e, a n d i n dustr ia l p ro gre ss c u rv e), w hic h i s d esc rib ed b y t h e f o llo w in g r e la tio nsh ip : E ach t im e t h e o utp ut q uantity d ouble s, t h e u nit l a bor h ours a re r e d uced a t a c o nsta nt r a te . F or e x am ple , a ssu m e th at a m an ufa ctu re r h as a n ew c o ntr a ct fo r 1 6 p ro to ty pe u nits a n d a to ta l o f 8 00 la b or h ours w ere re q uir e d fo r th e fir s t unit. P ast ex perie n ce has in dic ate d th at on sim ila r ty pes of units th e i m pro vem en t r a te w as 8 0 p erc en t. T his r e la tio nsh ip o f i m pro vem en t i n l a b or h ours i s s h ow n b elo w :

B y u sin g T ab le A 5.1 u nit v alu es, s im ila r la b or h ours p er u nit c an b e d ete rm in ed . L ookin g a cro ss th e 1 6 u nit l e v el a n d d ow n th e 8 0 p erc en t c o lu m n, w e f in d a r a tio o f .4 096. B y m ultip ly in g th is r a tio tim es th e la b or h ours f o r t h e f ir s t u nit, w e o bta in ed t h e p er u nit v alu e:

T hat i s , t h e 1 6th u nit s h ould r e q uir e c lo se t o 3 28 l a b or h ours , a ssu m in g a n 8 0 p erc en t i m pro vem en t r a tio . 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 3 9/4 4 P ag e 1 56 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . T A BLE A 5.1 L ea rn in g C urv es U nit V alu es 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 4 0/4 4 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 4 1/4 4 P ag e 1 57 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . O bvio usly , a p ro je ct m an ag er m ay n ee d m ore th an a s in gle u nit v alu e fo r e stim atin g th e tim e f o r s o m e w ork p ack ag es. T he c u m ula tiv e v alu es i n T ab le A 5.2 p ro vid e f a cto rs f o r c o m putin g th e c u m ula tiv e to ta l la b or h ours o f a ll u nits . I n th e p re v io us e x am ple , f o r th e f ir s t 1 6 u nits , th e to ta l l a b or h ours r e q uir e d w ould b e B y d iv id in g t h e t o ta l c u m ula tiv e h ours ( 7 ,1 36) b y t h e u nits , t h e a v era g e u nit l a b or h ours c an b e o bta in ed :

N ote h ow th e la b or h ours fo r th e 1 6th u nit (3 28) d if f e rs fro m th e a v era g e fo r a ll 1 6 u nits (4 46). T he p ro je ct m an ag er, k n ow in g th e a v era g e la b or c o sts a n d p ro cessin g c o sts , c o uld e stim ate th e to ta l p ro to ty pe c o sts . ( T he m ath em atic al d eriv atio n o f f a cto rs f o und i n T ab le s A 5.1 a n d A 5.2 c a n b e f o und in J e le n , F . C ., a n d J . H . B la ck , C ost a nd O ptim iz a tio n E ngin eerin g, 2 nd e d . ( N ew Y ork : M cG ra w -H ill, 1 983.) F O LLO W -O N C O NTR ACT E X AM PL E A ssu m e th e p ro je ct m an ag er g ets a f o llo w -o n o rd er o f 7 4 u nits ; h ow s h ould s h e e stim ate la b or h ours a n d c o st?

G oin g t o t h e c u m ula tiv e T ab le A 5.2 w e f in d a t t h e 8 0 p erc en t r a tio a n d 9 0 t o ta l u nits i n te rs e ctio n— a 3 0.3 5 r a tio . L ab or h ours fo r th e 9 0th u nit c an b e o bta in ed fro m T ab le A 5.1 : .2 34 9 × 8 00 = 1 87.9 la b or h ours . (F or ra tio s b etw een g iv en v alu es, s im ply e stim ate .) E xerc is e A 5.1 N orw eg ia n S ate llit e D ev elo p m en t C om pan y C ost E stim ate s f o r W orld S ate llit e T ele p hon e E xch an ge P ro je ct N SD C h as a c o ntr a ct to p ro duce e ig ht s a te llite s to s u pport a w orld w id e te le p hone s y ste m ( fo r A la sk a T ele co m , I n c.) t h at a llo w s i n div id uals t o u se a s in gle , p orta b le t e le p hone i n a n y l o catio n o n e arth t o c all i n a n d o ut. N SD C w ill d ev elo p a n d p ro duce th e e ig ht u nits . N SD C h as e stim ate d th at th e R & D c o sts w ill b e N OK (N orw eg ia n K ro ne) 1 2,0 00,0 00. M ate ria l c o sts a re e x pecte d to b e N OK 6 ,0 00,0 00. T hey h av e e stim ate d th e d esig n a n d p ro ductio n o f th e fir s t sa te llite w ill re q uir e 1 00,0 00 la b or h ours an d an 8 0 p erc en t im pro vem en t cu rv e is e x pecte d . S kille d l a b or c o st i s N OK 3 00 p er h our. D esir e d p ro fit f o r a ll p ro je cts i s 2 5 p erc en t o f t o ta l c o sts . 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 4 2/4 4 P ag e 1 58 P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d . T A BLE A 5.2 L ea rn in g C urv es C um ula tiv e V alu es 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 4 3/4 4 7 /2 2/2 017 U niv e rs it y o f P hoenix : P ro je ct M anagem ent: T he M anageria l P ro ce ss h ttp s://p hoenix .v it a ls o urc e .c o m /# /b ooks/1 259822338/c fi/ 6 /3 2!/4 /3 54/4 /2 @ 0:5 0.9 4 4/4 4 P ag e 1 59 A . P R IN TED B Y: l ttle m nta te @ em ail.p hoen ix .e d u. P rin tin g i s f o r p ers o nal, p riv ate u se o nly . N o p art o f t h is b ook m ay b e r e p ro duced o r t r a n sm itte d w ith out p ublis h er's p rio r p erm is sio n. V io la to rs w ill b e p ro se cu te d .