The Manual of Construction Productivity Measurement and Performance Evaluation_TA

· ", CONSTRUCTION INDUSTRY INSTITUTE THE MANUAL OF CONSTRUCTION i-PRODUCTIVITY MEASUREMENT 'AND PERFORMANCE EVALUATION

Views 287 Downloads 0 File size 13MB

Report DMCA / Copyright

DOWNLOAD FILE

Recommend stories

Citation preview

· ",

CONSTRUCTION INDUSTRY INSTITUTE

THE MANUAL OF CONSTRUCTION i-PRODUCTIVITY MEASUREMENT 'AND PERFORMANCE EVALUATION

,

.i

,.

The Pennsylvania State University H. Randolph Thomas Donald F. Kramer

\

'

I

\S

-

So",""

~,

Do""m~

May 1988

..

;

'

,.

( ~,~BLE

OF CCNTENI'S .

LIST OF FICJJRES

'J

vii i

LIST OF TABLE'S

PREFACE

CHAPI'ER

lX

1.

1

INI'ROax:TICN

Purpose Apptication Definitions Barriers to Monitoring Productivity Organization of Manual Use of Manual . . . .

1 1 2 3

4

CHAPI'ER 2.

6

ffil'MEWJRK Fa< PROax:TIVITY !"E'ISUREMENI' !'NO P~

7

EVAIIJATICN

Introduction . . . '. Characteristics of Costing Systems Usages . Costing 'System to Support Cost Control Costing Systems to Support Estilreting productivity Control . Track'cnly Important Activities Simplify the Code of Accounts Tailor the Level of Detail Simplify Units of ~asure Reduce the Amount of Data 'Sunmary . • . . .', . , . , ,

~'~

"1

~j

~,j

~:1

~~,

CHAPI'ER 3.

7 7 7 7

.'

9 9

15 15 15 17 17

17

:;)j' ....' ;:::

l

(

19

a:::N:EP'IUAL DESIrn OF THE PRCJlXJ:TIVITY MEA.SUREMENI' SYSTEM

19 19 19 20 21 21 21 24 24

Basis for System Developrrent Project Characteristics AsSLIIlptions . . , . . . System Criteria , , , . . . Framework for Productivity Measurement Reporting Requirements , . . , , Relationship to Other Accounting or Control Systems Selection of Activities to Monitor Developrrentof Productivity Codes Units of Measure Sunmary , .

~:i

(

5

P.ART I - ffil'MEWJRK Fa< PROax:TIVITY o:::NrROL

q ~,

i

26

28

.,

ji\1

{' ii

,'---

,f ,,~

~

( :...

o_~?:.. ..

1:'\

"

fij ~)

:::;

;:0:

;.•; :

.

o~

~':"

~::::;\

'0-'::

PART .II - BASIC o:::NC£PIS OF PROcu::I'IVITY MEASlJREMENr AND PERFORI-1AN::E EVAUJATIaI CHAPTER 4.

~1EASUREMENI'

OF QJANrITIES AND W)RJ->::1-

" rr v'~ ........

r

LA

canplicatsd, require a 3Up;:or: 5t.2.I: dedicated to rreasuring and quantiCJ'ing site activities, and result in an intolerable overhead expense [5 J• Alternatively, these contracLOrs rely on the lTUlch simpler accounting-type systallS such as the Ore shawn in Figures 3 and 4. Ho.iever, these provide little in the way of control. Fortunately, productivity rreasurarent can te greatly simplifie:l while still providing control. If canparisons are to te IT'dde with the project estiJM.te, then prcductivity com.rol lTUlSt originate fran the sane ccxie of accOllilts as 0Ces the cost con=l syst6Tl. Thereafter , productivity rreasurarent and control can i:e done separately fran the cost control syst6Tl. This separation allClO'S the freedan necessary to tailor the system to perfonn a specific furct.ion. ·The result is a systen that is simple, t.iirely, and ~ive. The re:nainder of the chapter develops this corcept through an explanation of the desired gcals rieeded to achieve the cost-effectiveness of the syst6Tl. i Track Only 1J+tpJrtant Activities

It is reccgnize:l that, on any given project, rrost of the work-hours are consurred by a SJM.1l number of activities. If one controls these· activities, he essentially controls nest of the project p=ople resources. The p:roblen i.rlhe=1t with cost control syst6TlS is that they are designect to track all project exp=nditures. This means that the contractor needs to ITEaSlire the outp.lt over the total project i:efore he can extract infonnation al::out the activities that are truly important. Additional overhead staff often needed to operate the systen and to interpret the results. Understandably, contractors who are not accustcrred to rrea.suring productivity are not eager to cOllilt light fixtures, d=rs, ladders, handrails, valves, pipe hangers, and the thousands of other minor itens that lTUlSt i:e installed. 3y tracking prcductivity separately, the contractor can choose to nonitor only those activities that he feels are important.

are

Simplify the Code of Accounts To rrany contractors, code of account structures like the one illustrated in Figure 2 are overly canplex. They require detailed narratives of what to include in each account, and considerable i:.iJre rray i:e needed to reconcile errors resulting fran costs t:eing chaIged to the wrong account. When tracking is done separately fran the cost control syst6Tl, field personnel·nee:l to deal only with the four-digit labor accounts as shown in Figure 5.

Tailor the Level of retail The level of detail in the code of accounts is typically established on the basis of costs and does not always fully suFP"rt the productivity control needs . 'I'M:> examples illustrate this pJint. The cost of 6-inch-diarreter stainless steel pipe is different han that of 2 112-inch-dianeter carton steel pipe. In the cost control systen, these ccmio::iities are tracked separately. BUt, fran the productivity rreasu.rarent viewpoint, the nay i:e nearly identical. Furthemore, a crew may install several types of pipe in a single day, thus requiring that the quantities and Io.Urk-hours be repJrted by type. Thus, the costing systan places an addeO b.rrden On a rep::>rting systan that contains rrore infonnation than may i:e required for productivity control.

=

15

c;.-

'v4J.~

l

".

(

c Activity That part of the code of a=ounts used to describe ;, phy5ica1 it"", of work or work task to be perfomecl. Ex2lIrple activities i.n::1ude piping, duct, backf~ll, roofing, and so forth. Activities are the primary labor accounts and are thus the productivity codes. (4 digits)

Figure 5.

Codes for Productivity Measurerrent

(

( 16 -;-',

....

j

:;

'._-

,

'

In the sec.~1d ex~le, the '.T.iC :::csc jifference of structural steel for ' comrerclal rrultistory and suuccural sLeel for warehouse-type buildings rk will be en g:llrg for an ext.a-rl2d per iOO of tirre

Q"eatsr .c'etail ard cbMJre irMJlve:l thcc\ je=tivity than sirrply sirrply estirretirg estirretirg h:w nu:n wxk p=t:"CEl1t carplete was d:re ird less exp;reive tren co.ntirg or rreasurirg tt"e t..nits =tplete::l

Best Slite::l wtere th=re

Easy to LEe 5iJ1ple to url=rstarrl

are ally a few itaTs· e3ch abtaS< is diff iOO t to ITEeSlre, ird the w:>rk rray last for ctivity 1b:lctivitieS or itaTs of w::Jrk sh:>..tl.d be of stnrt d.Jra tim Wxks best for a larg= ruTi:er of i taTs

.-----.-_.

,SiJlple

33

.-

tre

--

I.crg p:r ia:E rray elap;e

before en interns::liate milest:cre is rea:::tEd

~

be i.ra=..trate, e;p;cially if &.ere are few i taTs of if &.e activity d.tt'a"tim is lergthy

(

,

the three are executed simultaneously. The units completed.rrethcxl can be applied to concrete placement beCause the quantity is readily measurable. Another criterion for using the units corrpleted rrethod is that the t iJre needed to perform the corrpleted installation of an individual unit of output should be relatively shert, say a·day ·or less.' Trench excavation and the hanging of doors would qualify, whereas the testing of a piping system or the installation and alignrrent of equiprrent probably would not. Also, work involving multiple crafts is often not well suited to the units completed rrethod beCause it can seldom be corrpleted in a, day. The primary advantage of the units completed rrethod is that, when properly applied, it is toe most accurate anditherefore .reliable rrethod available .. It is a 'relatively objective rrethOd beCause it does not require a subjective opinion to determine what has been corrpleted. A third advantage yf this rrethod is that:an audit of the reported production is easily acconntJda~ The main problem with the units completed rrethod is the cost of data collection when the rrethod is improperly ppplied.

,

,

It is Worth noting that pipe fabrication '~nops can use this rrethod of measuring output bei::ause there is comronality !among many items. Since the fabrication shop is responsible for a relatively narrow scope of work (bulk quantity items), the breadth of the reporting 'system can be reduced without significant increases in manpower or cost requirements. Percent CClTplete (Supervisor CpinionJ Method A simple subjective approach'is to ask the supecvisor's opinion of the percentage of the task·which is completed. It is ~s~ful for relatively minor tasks, usually of short duration, where developrrent Jf a more complicated intermediate milestone or level of effort formula is not justified. Painting, dewatering, architectural trimming, and landscaping ,are candidates for this approach. Level of Effort

(

A control system requlnng the physical measurement of numerous items of work would too burdensare and costly. Q1e .way to simplify the rreasurement process is to assign a predetermined percent'complete.to a task on·the basis of the corrpletion of various subtasks. The percentage' is based upon the relative work-hours required to complete.each subtask. TO illustrate the level of effortrrethod, the follOwing example is considered in which a contractor must install 1,708 small-bore pipe.hangers. The following list of subtasks involved shO\olS the relative level of effort required for each.' The relative level of effort, or weighted completion status, is defined as rules of credit.

( 34

10,

.. ' .... ...•

; ;

"

1~g

SUbtask Fabricate Install Preservice iJ1sI;ection Total task

L'n.i. t of

Rules of

~sure

Credit

))~,( ,til, ~~ JJ~4

0.40 0.50 Q.JQ . l.00

each each each

each

·If, at sore point in tinE, 366 hange-rs have been fabricated, 185 installed, an:! 41 i.nspect.ed, then the cumulative number of hangers crnpleted is calculated as fo11=: Cumulative Quantity (each) = 366 (0.40) + 185 (0.50) + 41 (0.10) = 243.0

,I

'~

(1)

The subtasks are selected so that the status can be easily dete.Dni.ned by the . foreran, an:! 00 credit is given until the subtask is finished. The rules of credit rara.in the sane for all {tens of ;;ark within a given categJJ:Y or accOlli1t. In this example, they;;auld be the sane for all srrall-rore hangers, i.rresj::ective of the type or size of hanger. In another example, a contractor installs llOdular fornw:>rk for a reinforced corcrete basarent wall. First, the outside fonn is erected, braced, and aligned. The inside fonn is erected next. Thereafter, the foDTlS are braced, shcred, and plumbed as a unit. After the coocrete placment, the foDTlS are stripped, cleaned, and oiled .. Example rules of credit for this task are given bel=:·

=

Subtask Erect initial wall fonn Erect second wall fonn Final bracing and plumbing Strip and clean

Total task .

'.'

Unit of

Rules of

Measure

Credit

ft 2 ft 2 ft 2 ft 2 ft 2

0.90 0.70 0.10 0.10

The crew canoot take full credit for the ;;ark until after the fo= have teen raroved and cleaned. Notice that the sum of the rules does not add to l.00. This is because the first subtasks are applied to only half the total waLl

area.

=

A sare.lhat llDre detailed example involves the ere::tion of P~PJ.ng iscnetrics. Figure 10 shows the daily quantity report in which the foranan reports the status of four isaretrics. The rules of credit for pipe erection are presented bele"",,

35

( ~ail'i

, .... :..

·...;;'-'Pipe Erectcion 163X and 173X

'. foreman: Date:

':uantity Report

Page: Project: Off ice Code:

Smith

Status of WOrk: List quantity

i~talled

of

1

Ei609 2-0007

and check appropriate column 1, 2, or 3.

Erected - Count pipe erected in place but not bolted/welded cOlTpletely. (2) End Connections CClTPlete - Count pipe end connections cOlTpleted bolted/we,lded in place. (3) 'Trimmed and Complete - Count pipe cOlTpleted and ready for test.

(1)

Isometric Drawinq ~ A-267-30 A-267-32 0-:'27-43 0-527-43 0-819-02

I

I

I

I

Account ~ Spool 1732 Spool 1732 'Field Run 1735 Field Run 1735 Field Run 1735

Status of Work 1 2 I 3 X

\

X

I

X

I

I X

X X

I

I

X

I

X

I

\ Cunulative Q.lantity Installed Pipe Size to Date (in. ) (ft)

Total Q.lantitJ to be Installed (ft) 20

10

20

36

3

36

231

I

,1/2

200

231

II

1/2

147

i-1/2

10

10

X

I

(

I

I

(

I Figure 10.

Daily OJantity Report

,.

".

;(' \?:. ... l , .

'.

36

u

c.

~

,-~

::::>

> ~

".,. .

I

o

20

10

30

(

DAY

Figure 11.

~lication of Not Using Rules of Cre::lit, St:ru::tura1. Steel Erecti= (Account 05121)

( 38

--------'.

I

per piece increased by about 40 percent. It should be obvious that, for sane items, progress and productivity cannot be accurately measured without the application of rules of ccedit. Fortunately, rules of credit are not c~licated. Table 4 shoWs example rules for various carrrcdity .items. In general, no rrore than three to five subtasks should be considered. Each contractor develops a unique set of rules, which can remain unchanged from project to proJect. Criteria - The level·of effort· method should be used when the manner in which credit is awarded for partially c~leted work can lead to mIsleading interpretations of progress, productivity, and performance. From the previous examples, it should be readily apparent that the level of effort method is best suited for those activities that involve a.number of overlapping subtasks. These subtasks may include rrore than one craft, bot, ,obviously, each subtask must be measurable. As illustrated by the structural steel example, measurerent concepts can be sirrplified. To be used effectively, the status of subtasks must be easy to determine: e.g., a valve has be:J5 at all """lds All Eield Io.elds/rolte::l =tpleted ard a::o=pted Tested ard a::o=pte::l by

(

Q:

E1e::trical Terminatien

ea::h

70% termi.rated 30% acr:epted

Terrninatir'Xluctivity Calculations

Approach

Advantages

Disadvantages

Uses

Daily

Immediate feedback Provides an order of magni tude to a particular problem Supports the identification of causes

Wide variations possible which are difficult to explain Calculations done daily

Draws attention to problems that. occurred that day Facilitates the developnent of strategies to prevent cecccurrence

PeLied

Fewer calculations SlJlTTT1ar ies 'needed only pedodically Fluctuations in the 'data, not as great as with daily calculations

SlJlTTT1aries for Lack of timeliness of 'feedback upper-level managers Daily variations hidden Can be, useful 1n Limited number of data establishing points on which to base short-term conclusions regarding goals trends Fails to support the identification of cau-~

M:lving Average

Daily feedback Information not 'grossly distorted by one unusually good or bad day

Calculations tedious

Cumulative

M:lre closely relates to cost and profitability since total values are used

As 'MJrk-hours and

~re

quantity increase, the slope changes very slowly and by small increrents

Analysis of short-term trends

Forecasting probable outCOllE

critiquing overall progress

I I

~~

(

~:';

Daily Prcductivity Tre daily pro::iuctivity or the unit rate is defined simply as

~~f

Daily Pl::cductivity =

t:~

I~!:

Both w:>rk-hours and qu=tities must be chaIged to the sane account. In Figure 14, as an ~le, the electrician Cr:eN IooUrkEd 80 total =rk-h:lurs on that day, but only 19 of these charged to cable pulling. They cc:rnplete:i one pull of 175 feet. In this instance; the daily p:roductivity for cable pulling is calculated only for cc:rnplete:i =rk as

-=

. . 19 DailY "--'-. nc.uuctl.Vl.ty = 17 5 (Code 1830)

iji:::

:::;:,

l:~: . :jl~ :

( 3)

DailyQ.Jantity

E·,:

r:·

Daily Work-hours

= 0.109 ...::Jrk-hours/lin ft

(

Much infonnation can be gained by studying the daily productivity of a crew. SCIre causes of prcduetivity decline can be readily identified, whereas others are ItOre subtle and :require further investigation. To illustrate the usefulness of-daily pro::iuetivity statistics, three ~les are given using the activities ~Oped in Chapter 4.

Structural_S~ ~ igure 16 shOw'S a plot of the daily prcductiVl.ty of an i.rorn;orker crew erecting structural steel on a sUe-story office l::uilding. sUe days are·of particular interest, narrely, clays 4, 12, 15, 18, 21, ani 24. The poor pro::iuctivity on days 15 and 18 can be explained by cold ~atures, 20'P and 12'F, respectively. On day 'I, the weather was bad, ani the crew was sent hale after having ...::JrkEd for a1:xJut 3 hours. With this kn:>wledge, the mmager could better decide whether to 'lark or send the crew h:ne on subsequent clays. Day 12 is ItOre difficult to explain. The ~ature was 27'F, but the real problan may have been in the coordination of the =rk arrong the four subtasks. Much of the effort was spmt for plumbing, ani feM pieces of steel erected. On days 21 and 24, = h of the =rk involved bolting.

-=

(

Tre role of daily prcxluctivity data is to focus attention ani create awareness and sensitivity. Later in this chapter, it will be sh:rwn ~ other data can be used in conjW1ction with daily productivity to help isolate other

productivity problE!rlS. Pipe Erection - The daily productivity of a pipe erection CreM installing sUe-in::h-&.arreter pipe spools (account 1732) is sh:rwn in Figure 17. The CreM was ...::Jrking a 58-l-Dur/-el< schedule. During the first nine days, the daily crew productivity ...::Jrsened at a steady rate. Thereafter, e=atic variations are observed·. .Given this infonnation, a foreran or superintendent sh:Juld i.rnrestigate the causes. Perhaps =rk assignrrents or interferen:es with other crews· are the problE!rlS. cable Pulling - The analysis of .cable pulling (account 1830) is similar As can be seen in Figure 18, steady in::reases in unit rates occurred after day 118 .. to that of pipe erection.

( 52

'r.""."""."' ."., . . . · "'X. ".

}

"

8

~

• Difficulties in Plumbing Coordination o Bad Weather D. Boltlng Only

w

o

w 6

0-

0:: W 0C/) lJl

'"

0::

4

:::>

o

::I:

::.:: 0:::

l.

o 2-h ~

.~,.

0-1

o

I

I

I

I

I

I

I

5

I

I

I

10

I

I

I

I

I'

I

15

I

I

I

I

20

I

I

I

I

I

25

I

DAY

. figure 16.

Unit Productivity for StrucLural SLeel Erection (J\CCOUIiL 05121)

I

I

I

I 30

(

12

a::

oct UJ

-..J

s

ffia.

6

Z

en

(

a:: ::::l o

:I:

~

a::

o

3l:

2 O-Oh....,....-r--r-"1r-T"""'T""""T"""'T""".,..-r-r-""-'-"-,Srr-""-""""-2b 12 S 4

DAY

Figure 17.

(

Unit Productivity for Pipe Erection (A=ount 1732)

( 54

·

.j .

·.

.-'

0.5

8 u..

0::

0

I

I

::.:: 0::

0

~

o-i

Iii

o

I

I

5

i

I

j.

i

I

10

I

I

I

i

I

I

15

Iii

I

20

I

I

I

I

I

25

Iii

DAY

Figure 19.

Unit and M::lving Average Procluctivity for the First

12 Days of Structural Steel Erection (Account 05121)

I

I

30

;; ,'.'

(

,I: Table

NrITY RATES

So far, this chapter has concentrated on studying the different forms of productivity calculations. It follows, however, that, as suggested in Table 8; added insight can be gained by individually reviewing the work-hour and quantity trends and relating them to productivi:~' trends. Figures 23 and 24 show cumulative and daily woc~-hours and quantities, respectively, for the small-bore pipe erection acti':i.ty described at the beginning of the chapter (account 1735). In figure 22, the productivity steadily worsened throughout the activity duration. In Figure 23, both the cumulative and daily curves show that, through werkday 15, the number of work-hours assigned to the werk was very consistent. Relative to the quant,ities",. figure 24 sho.-ls a different picture. Four distinct phases are seen; eaG:h 'with successively lower daily output. Only phase 4 can be asS€Ciated'with the number of craftsmen assigned to the work. The study of quantities helps to develop a sensitivity that can be applied to many projects. For exarrple, figure 25 shows. the Cl.IDJlative quantity curves for three similar steel erection projects. Project A·has been developed throughout this chapter. Project B proceeded satisfactorily until a change in methods on day 13 forced the crew to integrate its werk with that of another. On Project C, the fact that steel deliveries occurred very slowly fran the beginning is clearly reflected. Where the level of effort method of measurement is used, it is often possible to review quantities installed for each subtask. figure 26 illustrates the usefulness of this type of analysis. The figure shows·the' cumulative pieces of steel that have been erected, bolted, plUlT'bed, and tightened. The completion rates should be reviewed concurrently with the unit and cumulative productivity curves, shown in Figures 16 and 20, respectively. In Figure 26, the relationship between plumbing and bolting is of particular

&3

( Table 8.

Productivity Summary for small-Bore PiPe (Account 1735)

~,TIisal 'M=e~ly

Q..1antit'f

Week

Weekly Cumulative lin ft lin ft Installed Installed

(1)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

18 19 20 21 22 23

351 350 232 587 438 453 532 557 500 453 450 532 326 467 449 334 255 418 377 243 262 513 521

12 )

351 701 933 1,520 1,958 2,411 2,943 3,500 4,000 4,453 4,903 5,435 5,761 6,228 6,677 7,011 7,266 7,684 8,061 8,304 8,566 9,079 9,600

Work-hours

Weekly Work-hours (3)

460 500 440 500 450 350 390 515 315 310 330 440 300 430 480 310 310 470 570 600 610 725 645

CumJlative Work-hours

(4) 460 960 1,400 1,900 2,350 2,700 3,090 3,605 3,920 4,230 4,560 5,000 5,300 5,730 6,210 6,=~O

6,'';0 7,2':0 7,8"70 8,470 9,080 9,805 10,450

Product i'l i ty Weekly. Cum..Jlati·/, work-hours! work~hour: lin ft lin Et

(5)

(6 )

1.31 1.43 1.90 0.85 1.03 0.77 0.73 0.92 0,63 0.68 0.73 0.83 0.92 0.92 1.07 0.93 1.22 1.12 1.51 2,47 2.33 1.41 1. 24~

.1. 31 1. 37 1.50 1.f 1.. 1.12 1.05 1.03 0,98 0,95 0.93 0,92 0,92 0.92 0,93 O. r O.

0,95 0.98 1.02 1.06 1.08

Gt• ,= ~

1.l~

c;A,,.~ . .. UP)......

CoI'1

( 64

)

/

._.

240 2000-1

_

CUMULATIVE WORl .10 f-

~

...J ::J

..,

~

u .07

::J

°

.------.... -.

."".-

14,000

28,000

42,000

CUMULATIVE· QUANTITI ES

Figure 32.

56,000 70,000 84,000 (LINEAR FOOT)

Productivity Forecast for Cable Pulling (Account 1830)

·c

,.;-;-

.\.

81

~

( ,.

This comparison shows "~at che graphical forecast can be more accurate than the analytical ones. The projected work-hours to caTl'lete can be readi ly calculated frem the productivity forecast. RELIABILITY OF ESTIMATES

In preparing the project estimate, managers and estimators make the best use of available data. Unfortunately, the time available for preparing the estimate is usually limited, and the inforrnationmay be incolTt'lete. As a result, quantity takecffs may be in error, or, in some cases, a work task may be completely overlooked. Productivity data is subject to wide variation, so the assumptions made in the estimating phase will seldom exactly coincide with reality. Accordingly, the estimate should not be considered a d=ument without fault, and, when variances develop during the control phase of a project, analysis of results should always consider estimating errors as a possible reason. ( To illustrate, it is assumed that the work-hours for .a given activity are projected to overrun. Some of the possible explanations for the overrun are listed below.

- The total quantity of work may have been underestimated, requiring rrore labor and material than budgeted. - 'The estimated work-hours per unit of work may have been unrealistically low. ~Crew productivity may have been lower than what should reasonably be expected for a var iety of reasons. Next, it .is assumed that the projected work-hours ocr another activity are expected to be right on target. This dces not necessarily mean that everything was correct. In fact, either of the following could be the situation: - A. low quanti·ty estimate may be offset by better-than-budgeted productivity. -' Worse-than-budgeted productivity may be compensated for by fewer quantities than budgeted.

( .

Thus, productivity control is much rrore than simp11 watching for negative total variances withm and arrong activities or a=ounts. The status of an account is the compcsite status of a number of parts, and each of those parts should be watched, even if productivity appears favorable.

This chapter has illustrated simple analytical and graphical techniques for cO!Tt'arlng actual performance to the control budget and for making . work-:-hour forecasts. Minimal information is required, and the output is in a conclse and easy-to-understand form.

(, . 82

.

,

f::

...... c. e.dity items. HQ'"IEver, the level of effort rrethod can also be used to monitor ~lex, one-of,..a-kind items, as is illustrated by the exanple for the erection of an absorber tower on a process plant.

r:"

For this work, it is estimated that 520 tons of structural steel are involved. Table 12 shows the status of the work at an early stage of completion. Different units of measure are used f~c the various subtasks. Several items, such as shakeout, plumb, and'punchl:3t, are associated with the entire absorber tower and cannot be conveniently r,=cated to a particular tier or bay. These subtasks are measured on the basis c,: percent ccxt;>lete. Girts and sag rods are measured according to the number ·of bays finished. The columns, beams, braces, and connections are rreasured by the piece. The overall progress of the erection operation is determined by multiplying the percent ccxt;>lete of each subtask by the respective rule of credit. The table shows that the work is 15.6 percent ccmplete. The equivalent tonnage erected is 15.6 percent of the 520 tons, or 80.5 tons.

L.

;

'.

}

f~",

l I

A characteristic problem with construction productivity is that it is continually changing. This variability can result in significant analysis and forecasting inaCcuracies when using the linear approach defined by Equation 7 in Chapter 6. The need to develop accurate forecasts is particularly acute during the early stages of an activity. To overcome the problem of variations in cumulative unit rates, some contractors have developed standard productivity performance curves for key· comrodity items. The curves reflect the contractor's expectations, on the basis of his experience, of how productivity performance should change during the.installation process. An example of a standard productivity performance curve for small-bore pipe installation (account 1735) is shown in Figure 33.

84

f-~::~;r.

" ....

.:.;.>

'.-. -,0>:.

-,.,

[:::.~:..'.,;;

\.0>•. ,.1

..

,