Method for Calculating Schedule Delay

Considering Lost Productivity

Hyun-Soo Lee, M.ASCE1; Han-Guk Ryu2; Jung-Ho Yu3; and Jae-Jun Kim, M.ASCE4

Abstract: A delay claim often occurs when a difference between the actual completion date and the contract completion date exists. The
duration of a delay is an essential piece of information required for determining the cause of a delay. However, it is difficult to analyze
a delay claim due to the fact that numerous factors that cause this delay, thereby making it a very complex issue. One of such factors is
the lost productivity or loss of productivity. Despite the fact that it is one of the major causes of delay, there have been only a few studies
that focus on converting lost productivity into delay duration carried out to date. Claims for productivity losses are generally the result of
tension between the contractor and the owner. This tension arises due to the great difficulty involved in quantifying disruption effects.
Thus, to calculate accurately the delay duration, a logical method for analyzing schedule delay caused by lost productivity is necessary.
Therefore, in this study, we propose a method for analyzing construction schedule delay where this lost productivity is taken into
consideration. This methodology was implemented on a case project to ascertain its practicability, and to decide whether it can be utilized
in the case of a delay claim related to lost productivity. The significance of this paper is twofold. One is the method to convert the lost
productivity into the delay duration, which can be applied to reasonable delay claim settlement. The other is the process to analyze the
construction schedule delay considering lost productivity.
DOI: 10.1061/共ASCE兲0733-9364共2005兲131:11共1147兲
CE Database subject headings: Claims; Delay time; Productivity; Data analysis; Databases; Construction industry.

Introduction ties 共Bubshait and Cunningham 1998兲. Thus, the ascertainment

of the period of project delay serves as basic information for the
The duration of a construction project is an important factor to set apportionment of responsibility, which may be a highly complex
forth when entering into a construction agreement. If a contractor operation in cases with concurrent causes 共Shi et al. 2001兲. As-
works within a planned parameter, he/she should be able to finish signing responsibility for project delays is critical to the allocation
the construction project in a timely manner. However, compared of responsibility for time-related costs 共Al-Saggaf 1998兲. In this
to other industries, it is difficult to complete a construction project respect, when a delay claim occurs, it is very important to assign
in which many construction trades participate and numerous un- responsibility and magnitude to the delay. However, many
known variables exist. When such difficulties arise, construction sources and causes of construction delays exist, and it is often
schedules are delayed, and consequently delay claims occur. difficult to analyze the ultimate liability in delay claims 共Kraiem
Delays in construction may be caused by the owner, the con- and Diekmann 1987兲. Lost productivity or loss of productivity is
tractor, acts of God, or a third party. They may occur early or late one of the most important causes of delay among the various
in the job, alone, or with other delays. In whatever cases, negoti- causes of construction delays. Thus calculating the effect of lost
ating a fair and timely damage settlement is beneficial to all par- productivity on delays is an intricate issue.
Although several studies have been directed toward the issue
1
Associate Professor, Dept. of Architecture, Seoul National Univ., of analyzing delay and lost productivity 共Kallo 1996; Al-Saggaf
Seoul 151-742, Korea 共corresponding author兲. E-mail: 1998; Bubshait and Cunningham 1998; Finke 1998; 1999; Kar-
[email protected] tam 1999; Reichard and Norwood 2001兲, but they have been
2
PhD Candidate, Dept. of Architecture, Seoul National Univ., Seoul
mostly concerned about converting lost productivity into cost.
151-742, Korea. E-mail: [email protected]
3
PhD, Full–time Lecturer, Dept. of Architectural Engineering, There have been only a few studies conducted on concrete meth-
Kwangwoon Univ., Seoul 139-701, Korea. E-mail: ods for converting lost productivity into delay duration.
[email protected]; formerly, Dept. of Architecture, Seoul National The purpose of this study is to propose a practical method for
Univ., Seoul 151-742, Korea. converting lost productivity into delay duration. Several concepts
4
Associate Professor, College of Architecture, Hanyang Univ., Seoul pertinent to the lost productivity and its consequent delay analysis
133-791, Korea. E-mail: [email protected] are introduced. The method is presented and implemented to a
Note. Discussion open until April 1, 2006. Separate discussions must case project to ascertain its practicability.
be submitted for individual papers. To extend the closing date by one Among the various factors that cause delays, this study fo-
month, a written request must be filed with the ASCE Managing Editor.
cuses on the factors that cause loss of productivity. There are a
The manuscript for this paper was submitted for review and possible
publication on May 15, 2003; approved on March 22, 2005. This paper is few kinds of productivity, e.g., labor productivity, equipment pro-
part of the Journal of Construction Engineering and Management, Vol. ductivity, etc. This study has focused on the labor productivity
131, No. 11, November 1, 2005. ©ASCE, ISSN 0733-9364/2005/11- because labor productivity representatively shows all kinds of
1147–1154/$25.00. productivity. The effect of lost productivity on delay duration will

JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / NOVEMBER 2005 / 1147

 Basic Concepts for Considering Lost Productivity
Productivity may be defined as the quantity of work produced per
man hour, equipment hour, or crew hour 共Finke 1998兲. As shown
in Fig. 1, the lost productivity or loss of productivity is repre-
sented by the productivity impacted by unexpected factors or im-
pact factors. For example, a curtain wall crew consisting of five
workers installing 34.65 m2 / h has a productivity rate of
6.93 m2 / man h under ideal conditions without any other impact
factor. However, if the work is impacted by another impact factor
such as unexpected adverse weather, the work productivity will
decline, and unexpected extra time will be required for the im-
pacted work productivity to be unimpacted work productivity.
To introduce the basic concepts regarding delay and produc-
tivity, it is necessary to define the terminologies. Planned work
duration 共D P兲 is the work duration with the planned work produc-
tivity 共P P兲 of as-planned schedule. Actual work duration 共DA兲 is
Fig. 1. Lost productivity the work duration with the average productivity of as-built sched-
ule 共PA兲. As-built schedule is widely accepted for determining the
impact of project delays. It is basically a comparison of what was
planned to what actually occurred at the job site 共Bramble and
Callahan 1992兲. Start time variance 共VS兲 is the difference between
be calculated against critical works, which is represented as a line
the actual start time of work and the finish time of the preceding
in the LOB form of a project schedule. The LOB form of a project
work of the as-built schedule. Finish time variance 共VF兲 as the
schedule is practical in understanding the changes of productivity
difference between DA and D P is composed of DN and D P.
visually.
When lost productivity factors occur, the lost productivity 共PL兲
of work can be calculated using Eq. 共1兲, where PU and PI denote
the unimpacted productivity of the activity and the impacted pro-
Schedule Delay Calculation Method Considering ductivity of a given activity, respectively. If the work is planned
Lost Productivity properly, then the PU will be the same as the P P

Review of Schedule Delay and Lost Productivity LP = Pu − Pi 共1兲

Calculation Methods
The lost productivity entails lost work, the amount of which could
Many studies or methodologies directed at analyzing delay and be completed without the lost productivity. The quantity of lost
lost productivity have been reported. The current status of delay work due to the lost productivity 共QPL兲 can be calculated by the
calculating studies related to lost productivity can be summarized following equation, where LI denotes the daily average labor dur-
in three cases as follows. 共1兲 It is assumed that lost productivity ing the impacted work duration 共DI兲
claim is a different type of impact claim than a delay claim.
Claimants frequently confuse these two types of claims because QPL = 共PU − PI兲 ⫻ LI ⫻ DI 共2兲
the impacts of both occur simultaneously 共Bramble et al. 1990兲.
共2兲 Lost productivity claims are limited to studies of converting The lost productivity also entails lost duration DPL, which as an
lost productivity into cost such as total labor cost method, mea- opportunity duration could be worked as much as QPL with the LI
sured mile analysis, expert estimate, industry factors, historical and PU. The DPL can be calculated by the following equation. All
productivity data, etc. 共Kallo 1996; Reichard and Norwood 2001兲. the variables in the equations hereafter are summarized in the
For example, the measured mile analysis is conducted by com- Notation
paring productivity during an unimpacted period of time with the
productivity during an impacted period of time. The claim
amount is the difference between the two productivity measures
multiplied by the costs associated with the lost productivity. Dam-
DPL =
QPL 共PU − PI兲 ⫻ LI ⫻ DI
LI PU
=
LI PU
= DI 1 − 冉
PI
PU
冊 共3兲

ages by lost productivity are calculated by subtracting the esti- When the contractor’s claim includes the “ripple effect,” that is a
mated costs from the actual costs incurred. Because the actual request for compensation for activities whose productivity suf-
costs include the increased labor costs from the lost productivity, fered indirectly due to the owner’s actions, the situation is further
the cumulative effect of changes is addressed in the total cost complicated. The owner is usually reluctant to accept the exis-
overrun. 共3兲 Delay causes are conceived as activities in a project tence of this ripple effect because it is not readily seen and be-
schedule such as a method of “what-if” evaluation or “but-for” cause it may be used to cover up the inefficiency caused by the
schedule 共Al-Saggaf 1998; Bubshait and Cunningham 1998; contractor’s mismanagement 共Abdul-Malak et al. 2002兲. There-
Finke 1998, 1999, Kartam 1999兲. “But-for” schedule results from fore, the characteristics of the impact factor must be known to
“pulling out” all owner delays that affected the as-built critical fairly assign the DPL.
path. The amount of compensable delay is the difference in time Impact factors affecting the DPL are categorized according to
between the actual completion date on the as-built schedule and independence and impact on the next impact factor. An indepen-
the completion date on the “but-for” schedule. dent factor means that an impact factor does not have any effect
As a result, a study concerning the methods of calculating on the duration of subsequent works 共see Scenario 1 of Fig. 2兲.
schedule delay of lost productivity is not sufficient. An impacting factor means that an impact factor adversely affects

1148 / JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / NOVEMBER 2005

 Fig. 2. Lost productivity factors’ characteristics

the duration of subsequent works when the next impact factor

occurs even before the previous impact factor’s ripple effect has
finished 共see Scenario 2 of Figs. 2 and 3兲.
As shown in Fig. 3, if the PL共i−1兲共F j−1兲 is impacted by F j,
the lost productivity will consist of the PL共i−1兲共F j−1兲␣, the
new PLi共F j兲, and the intersection of PL共i−1兲共F j兲 and PLi共F j兲, or
PL共i−1兲共F j−1兲␤, where PLi共F j兲 denotes the portion of ith productiv-
ity due to the ith factor. Usually the PL共i−1兲共F j兲 is not equal to Fig. 4. Calculation process for schedule delay
PL共i−1兲共F j−1兲 because PL共i−1兲共F j−1兲 would naturally change through
a mode of learning curve as times passes. Thus the PL共i−1兲共F j兲,
which is PL共i−1兲共F j−1兲 impacted by F j, can be written as countable for the work delay, and to calculate how much is im-
PL共i−1兲共F j−1兲␣. The ␣ is defined as the self ratio, the extent which pacted. The analysis of the delay duration can be carried out in
PL共i−1兲共F j−1兲 would naturally change from the initial point of F j−1 the procedure depicted in Fig. 4.
to the initial point of F j without the F j. The ␤ is defined to explain 1. Propriety analysis of as-planned schedule and as-built sched-
the intersection as the impacting ratio, the extent which ule. In this phase, propriety is analyzed to verify the reason-
PL共i−1兲共F j−1兲 affects PLi共F j兲. ableness of the as-planned schedule and D P considering the
average labor productivity, appropriate resource allocation,
labor usability, weather, appropriate materials, machine sup-
Process for Schedule Delay Analysis ply, and so forth. After verifying that the as-planned schedule
A common method for calculating a schedule delay is by compar- and D P are reasonable, the as-built schedule and DA are
ing the as-planned schedule and the as-built schedule prepared by examined based on the evidence 共e.g., detailed work sched-
the critical path method 共CPM兲 共Kraiem and Diekmann 1987; ules, updated schedule, daily reports, correspondence, delay
Bubshait and Cunningham 1998兲. The critical works in the as- description, etc.兲. The as-built schedule reflects the actual
built schedule ultimately impact the delay duration. The critical progression of events that occurs during the execution of
works can be classified into two types. One type is the work the project. As a result, the as-built schedule should be made
impacting on the project completion date and the other type is the by carefully studying the project reports and documents
work that has no impact on the compeltion date. It is therefore 共Kraiem and Diekmann 1987兲 and based on this evidence, it
necessary to analyze which works influence the completion date can be justified whether the as-built schedule and DA were
and to determine the degree of their impact. In other words, it is appropriate.
necessary to examine what the cause is, to determine who is ac- 2. Analysis of critical works in as-built schedule. A construction
project normally proceeds at a pace that is usually different
from the as-planned schedule. The criticality of individual
activities in a CPM network changes due to delays and ac-
celerations in construction 共Arditi and Robinson 1995; Shi et
al. 2001兲. A project delay is the accumulated effect of delays
in individual activities 共Shi et al. 2001兲. Because the delay of
critical work in the as-built schedule ultimately affects the
project completion date, delay causes within the critical
works must be recorded and analyzed on the as-built sched-
ule. It is important to know which work among the critical
works influences the project delay and the extent of their
influence on the as-built schedule. The extent of the impact
on the project delay should be analyzed comprehensively by
Fig. 3. Change of PL共i−1兲共F j兲 impacted by F j considering the project characteristics, field environment, the

JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / NOVEMBER 2005 / 1149

 impacted work characteristic, the completed work extent, and
so forth.
3. Analysis of VS and VF. Through a comparison of the critical
works in the as-built schedule and the works in the as-
planned schedule, VS and VF can be calculated. After cal-
culating VS and VF, the causes or impact factors of the VS
and VF must be found and analyzed by investigating the
evidence.
4. Finding the evidence of VF. Once the claimant has been con-
vinced that the construction could be completed if the works
were carried out according to the as-planned schedule and
that the as-planned schedule is sufficient to meet the work
sequencing restraints, then he/she will question why the Fig. 6. DW including only DPL
works were actually delayed. A construction claim is an as-
sertion of and a demand for compensation by way of evi-
dence produced and arguments advanced by a party in sup- Therefore, contractor’s difference 共DD兲 between DC and D P
port of its case 共Kululanga et al. 2001兲. So, the DN of VF can is a mistake made by the contractor as a result of a miscal-
be found and assigned to the owner, the project contractor, or culation of the work duration in planning. DD can be calcu-
a third party according to their responsibility. If a DA exclud- lated by the following equation:
ing the DN is equal to the D P, the delay calculation is com-
plete. In this case, the actual productivity of the work is QP
DD = DC − D P = − DP 共5兲
consistent with the P P. However, if the DA is not equal to the LU PU
D P, the calculating process proceeds to the next phase. If the
Work delay 共DW兲 consists of DD, DPL, and ␧ as shown in the
project contractor has no evidence to prove the PL, the cal-
following equation, where DD and DPL are independent vari-
culating process should also be stopped. In other words, if ␧
ables and ␧ is an extraneous variable that accounts for any
is not verified, then ⑀ is the contractor’s responsibility 关See
delays other than DC and DPL
Eq. 共6兲兴.
5. Analysis of delay causes. After the VF evidence is confirmed, DW = DD + DPL + ␧ 共6兲
the impact factors can be analyzed by comparing the DA with
the D P. The contractor can finish planned work quantity 共Q P兲 DW are classified into three cases: DW including DD and
with the unimpacted daily labors 共LU兲 and the PU within the DPL , DW including only DPL, and DW including DPL less DD.
duration 共DC兲. This can be calculated by the following equa- 1. DW including DD and DPL. DW includes all DD and DPL
tion, where QU denotes the quantities worked in the normal during the impacted work duration as shown in Fig. 5.
and realistic work conditions of an unimpacted work dura- When DD is greater than zero, as shown in the following
tion 共DU兲, and QI denotes the quantities worked in the im- equation, the delay is caused by the contractor’s mistake
pacted work duration 共DI兲 such as the allocation of lower labor and productivities
than the as-planned schedule. In this case, the contractor
QP QU + QI could not be compensated or, rather, should compensate
DC = = 共4兲 the owner for the liquidated damage
LU PU LU PU
Time is of the essence in a construction contract. Typically, a DD = DC − D P ⬎ 0 共7兲
time period is defined as the contract duration. The contrac- 2. DW including only DPL. The DW includes only DPL in the
tor is obliged under the contract to achieve substantial case where the DA of the as-built schedule is equal to the
completion within the specified period 共Shi et al. 2001兲. D P of the as-planned schedule as shown in Fig. 6. When
DD is equal to zero, as shown in the following equation,
the delay is caused only by PL. As P P is equal to PA, the
contractor shall not be held accountable for the delay

DD = DC − D P = 0 共8兲

Fig. 5. DW including DD and DPL Fig. 7. DW including DPL less DD

1150 / JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / NOVEMBER 2005

 of the gondola. In this case, the lost productivities caused by the
change order and the machine malfunction are different according
to single impact.
The combined impact is the combination of integrated impact
and single impact. For example, if 1 week later after the design
change and the material items change of the main entrance oc-
curred at the same time and at the same space 共integrated impact
factors兲, and there was also a malfunction in the gondola 共single
impact factor兲, the lost productivities of the integrated impact
factors and single impact factor should both be calculated.
Fig. 8. Integrated, single, and combined impact
Sum of PL„SPL…
After analyzing the characteristics of impact factors by using the
3. DW including DPL less DD. When DD is less than zero, as time-space 2 ⫻ 2 matrix, the sum of PL共SPL兲 can be quantified
shown in the following equation, DW is calculated by only according to the impact factor’s characteristics. PLi changes ac-
DPL. If it were not for the DPL, the actual work could cording to the characteristics of the impact factors from F j to Fn.
have been finished earlier than D P, as demonstrated in The SPL from Fi to Fn can be written as the following equation:
Fig. 7. If the DC is shorter than the D P, the work might be n
performed by acceleration or the use of more labor than
was originally planned, but the effort cannot be recog-
SPL1_共F1,F2, ¯ ,Fn兲 = 兺
i=1
PL1共Fi兲

nized as a compensable delay

n

DD = DC − D P ⬍ 0 共9兲
SPL2_共F2,F3, ¯ ,Fn兲 = 兺
i=2
PL2共Fi兲

¯ ¯
Assignment of Lost Productivity Duration n

Calculation of Lost Productivity

SPLk_共Fk,Fk+1, ¯ ,Fn兲 = 兺
i=k
PLk共Fi兲

After DPL is verified, the DPL must be assigned to the source by

¯ ¯
analyzing the impact factors. The responsibility for delays can fall
on the owner, the contractor, or third parties. Compensation or n
liquidated damage can be determined in proportion to the extent
of the impact: self ratio 共␣兲 and impacting ratio 共␤兲. Impact fac-
SPL共n−1兲共Fn−1,Fn兲 = 兺
i=n−1
PL共n−1兲共Fi兲 共10兲
tors with ␣ and impact factors with ␤ can affect the productivity
as a manner of integrated impact, single impact, and combined
impact according to the time-space 2 ⫻ 2 matrix, as shown in Sum Assignment of DPL
Fig. 8.
When given impact factors affect the work productivity at the It is essential that distinctions of excusable/nonexcusable delays
same time and at the same space, the extent of the integrated are made when analyzing delays. Excusable/compensable delays
impact represents the extent to which proceeding impact factors are due to some actions or omission of the owner, for example,
equally affect the productivity of subsequent work. Integrated im- lack of site access, or late arrival of owner-furnished material or
pact factor is analogous to the characteristics of concurrent de-
lays. Concurrent delays are used to describe two or more delays
that occur at the same time by one or more of the parties 共Richter
1983; Kraiem and Diekmann 1987; Bramble and Callahan 1992;
Rubin et al. 1992兲. For example, when the curtain wall design
change of the main entrance and the material items’ change of the
curtain wall of the main entrance occurred by the owner at the
same time, this would cause a delay in the work. In this case, the
lost productivities caused by the design change and the material
items’ change of the main entrance occurred integrally according
to integrated impact factors.
When given impact factors affect work productivity at a dif-
ferent time and at a different space, the extent of the single impact
represents the extent which proceeding impact factors have dif-
ferent effects on the productivity of the subsequent work. For
example, when a change order concerning the curtain wall work
on the fifth floor is made, this would cause a delay in the work.
Then, several days later after the change orders, the curtain wall
work on the seventh floor was delayed because of a malfunction Fig. 9. Master plan of project

JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / NOVEMBER 2005 / 1151

 Fig. 11. ACT27 schedule

Korea. The construction project was performed in the manner of a

traditional delivery system or a single contract. One of the writers
Fig. 10. Comparison of as-planned schedule and as-built schedule was actually involved in the project. Productivity data were col-
lected daily in a database and used for the delay analysis in this
case study. The following sections will explain the analysis con-
equipment. In such cases, the contractor would be entitled to ducted on the delay by using the process proposed in this paper.
damages for extra costs incurred unless there is an enforceable
contract clause barring such recovery. Excusable/noncompensable
Propriety Analysis of As-Planned Schedule and
delays are delays for which neither party is at fault such as: acts
As-Built Schedule
of God, epidemics, etc., as set forth in the delay clause. Time
extension is the only remedy for such delays. Nonexcusable de- Common work duration in the as-planned schedule was investi-
lays are delays caused by the contractor. These could include gated thoroughly and the as-built schedule was reasonably
failure in coordinating the work, insufficient manpower on the compiled according to the evidence that was stored in the data-
job, late delivery of equipment furnished by the contractor, low base. Therefore, the as-planned schedule and as-built schedule
productivity, defective work that must be promptly corrected, etc. can be seen to be of acceptable level to be used in the proposed
Such delays could be compensable to the owner in the form of methodology.
liquidated or actual damages 共Rubin et al. 1992兲.
After calculating PL of each impact factor and SPL, the respon-
sibility for the SPL can be assigned to the owner, the contractor, or Analysis of Critical Works in As-Built Schedule
the third party, based on the causes. The sum of the DPL can be
The planned completion date of the master plan was October 31,
calculated as the following equation:
1999 but the actual completion date was November 25, 1999. The
SPLi共F j兲 ⫻ LI ⫻ DI project was delayed for 25 days 共See Figs. 9 and 10兲.
SDPLi共F j兲 = 共11兲 The planned critical works are from ACT1 to ACT16 共the
LI PU
planned project duration is 930 days兲 but the actual critical works
are from ACT1 to ACT4 and from ACT17 to ACT29 共the actual
project duration was 955 days兲. Thus, from ACT1 to ACT4 and
Case Study from ACT17 to ACT29 were the impacting works of the actual
delay of the project completion in this case study. This case study
The case project for this case study was the construction of a focuses particularly on the ACT27 共the exterior curtain wall ac-
20-story office building located in the business center of Seoul, tivity兲 which was delayed the longest.

Table 1. Productivity Data Summary of ACT27

Work Average work
quantity productivity Impacted Impact
Duration Work area Manpower 共m2兲 共m2 / man day兲 duration factor
June 23–June 24 Gondola installation 40 — — — —
June 25–July 8 2 step of 8th floor–6 step of 12th floor 249 1,725.75 6.93 — —
July 9–July 10 Material delivery 19 — — — July 9-first change order
July 11–July 27 1 step of 13th floor–5 step of 19th floor 456.5 2,880.35 6.31 * —
July 28–July 29 Material delivery 9 — — — July 29–second change order
July 30–July 31 6 step of 19th floor–4 step of 20th floor 52 221.62 4.26 * —
August 1–August 3 Rainfall — — — — Adverse weather
August 4–August 6 4 step of 20th floor–7 step of 20th floor 80 272.57 3.4 * —
August 7–August 8 Parapet of roof 44 280.46 6.3 — —
August 9–September 5 Finishing and gondola uninstallation 258 1,726.87 6.69 — —

1152 / JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / NOVEMBER 2005

Table 2. SDPL of ACT27 Table 3. Variables Used in Equations
Classification Q共Fi兲 P共Fi兲 SPL共Fi兲 L共Fi兲 D共Fi兲 QPL共Fi兲 SDPL共Fi兲 Daily Work
Work Work average duration
Unimpacted 1,725.75 6.93 — 17.8 14 — —
Work status quantity productivity labors 共days兲
duration
First 2,880.35 6.31 0.62 28.5 16 282.72 1.43 Planned work Q0 PP LP DP
impacted As-built work QA PA LA DA
duration Work of unimpacted duration QU PU LU DU
Second 221.62 4.26 2.67 26 2 138.84 1.83 Work of impacted duration QI PI LI DI
impacted
duration
Third 272.57 3.4 3.53 26.7 3 282.75 4.67 3
impacted
duration
SPL2共IF2,IF3兲 = 兺
i=2
PL2共FLi兲 = PL2共F2兲 = 6.93 − 4.26 = 2.67

SLP3_IF3 = LP3_IF3 = 6.93 − 3.4 = 3.53

Analysis of VS and VF
The D P of ACT27 was from June 17, 1999 to August 15, 1999
and the DA of ACT27 was from June 23, 1999 to September 5, SDPL Calculation of First, Second, and Third F
1999 as shown in Fig. 11. It was proved that the 6 days of VS was SPL1共F1兲 is 0.62共m2 / man day兲 , L1 is 28.5 共labor days兲, and D1,
the cause of the proceeding work ACT26 performed by the same except for a DN 共July 22兲 due to rain, is 16 days. QPL1 and SDPL1
subcontractor and the VF is 15 days. are calculated as follows 关See Eqs. 共2兲 and 共11兲兴. Other SDPLs are
calculated in the same way as shown in Table 2.
Finding Evidence of VF
QPL1 = 共PU − P1兲 ⫻ L1 ⫻ D1 = 0.62 ⫻ 28.5 ⫻ 16 = 282.72 m2
The lost productivity of the ACT27 was impacted by three impact
共12兲
factors 共FS兲 : the owner’s two change orders and adverse weather.
The productivity data of the ACT27 is summarized in Table 1
based on the construction productivity database. QPL1 282.72
SDPL1 = = = 1.43 days 共13兲
L1 PU 28.5 ⫻ 6.93
Cause Analysis of Work Delay
The total number of labor days was 249, the work quantity was Assignment of SDPL
1 , 725.75 m2, and no impact factor was found from June 25, 1999 According to the responsibility for the SDPL, the SDPL can be
to July 8, 1999. Thus LU are 17.8 labor days and PU are assigned to the owner, the contractor, or the third party. The first
6.93 m2 / man day and second change orders issued by the owner are excusable/
compensable delay and the adverse weather is excusable/
249 men
LU = = 17.8 man/day noncompensable delay:
14 days • nonexcusable delay= 75− 共60+ 7.67+ 3.26兲 = 4.07 days,
• excusable/ compensable delay= 1.43+ 1.83= 3.26 days, and
1,725.75 m2 • excusable/ noncompensable delay= 3 + 4.67= 7.67 days.
PU = = 6.93 m2/man day From this, we can conclude that the contractor has the respon-
17.8 man/day ⫻ 14 days
sibility for the 4.07 days among the total 15 delays of ACT27
The DD can be calculated using Eq. 共5兲. The total work quantity 共more accurately writing FTV of ACT27兲 of the liquidated dam-
of curtain wall less parapet work quantity 共Q P兲 is 6 , 827.16 m2. age 共see Table 3兲.
The D P except for the DN is 54 days

Q0 6,827.16 Conclusion
DD = DC − D P = − DP = − 54 = 1.34 ⬎ 0
LU PU 17.8 ⫻ 6.93
Thus, DW includes DD and DPL during DI. An analysis of delays that occurred in a construction process and
the allocation of the responsibilities for the delays always entail
differences of opinion on the cause of the delay and on who
Characteristics of Fs and Calculation of SPL should be held accountable. In particular, when delays caused by
Because Fs occurred at a different time and at a different space, lost productivity are involved, the analysis becomes very compli-
the Fs represent single impact factors. The consequence of ana- cated. There are several reasons that contribute to the delay of a
lyzing the characteristics of the first, second, and third impact project. Many studies or methodologies for analyzing the delay
factor is independency 共␣ = 0 % , ␤ = 0 % 兲 in which PL共i−1兲共F j−1兲 have focused on these reasons. The reasons for a delay are usually
was dismissed during the following work duration. Thus, SPLi conceived of activities in a project schedule and the impacted
caused by F1, F2, and F3 can be calculated as follows: activities are analyzed without considering the impacted produc-
tivity. However, if some variables impact the next sequence of the
3 work in the construction project, the impacted work may become
SPL1_共F1,F2,F3兲 = 兺
i=1
PL1共Fi兲 = PL1共F1兲 = 6.93 − 6.31 = 0.62 lost productivity work. As only a few studies have been reported
on converting lost productivity into a delay, this paper presents a

JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / NOVEMBER 2005 / 1153

method for calculating schedule delays considering lost produc- PLi共F j兲 ⫽ portion of ith lost productivity due to jth factor;
tivity in an attempt to settle claims without litigation. The main QPL ⫽ lost productivity quantity;
conclusions from this study are as follows: VF ⫽ finish time variance;
1. The methodology introduced several concepts regarding VS ⫽ start time variance; and
delay and productivity, such as planned and actual work ␧i ⫽ portion of ith lost productivity not due to factors
duration, impact factors, lost productivity, the duration of detected.
lost productivity, start time variance, and finish time vari-
ance, etc.
2. Based on those concepts, a delay analysis process and References
equations for calculating the required values are developed
with which schedule delays can be analyzed with greater Abdul-Malak, M. A., Ei-Saadi, M. H., and Abou-Zeid, M. 共2002兲. “Pro-
accuracy. cess model for administrating construction claims.” J. Manage. Eng.,
3. The methodology was presented and implemented to a case 18共2兲, 84–94.
project to show its practicality. The case study indicates that Al-Saggaf, H. A. 共1998兲. “The five commandments of construction
the method is a more logical process for analyzing of the project delay analysis.” Cost Eng., 40共4兲, 37–41.
complicated delay situations, and thus can provide more de- Arditi, D., and Robinson, M. A. 共1995兲. “Concurrent delays in construc-
tailed analysis results on schedule delays. tion litigation.” Cost Eng., 37共7兲, 20–30.
We could also conclude that a properly designed database Bramble, B. B., and Callahan, M. T. 共1992兲. Construction delay claims,
could aid in the accumulation of statistical data on self ratio 共␣兲 2nd Ed., Wiley, New York.
and impact ratio 共␤兲, and contribute in the application of the Bramble, B. B., D’Onfrio, M. F., and Stetson, J. B. 共1990兲. Avoiding &
proposed methodology. As a result, further research on the utili- resolving construction claims., R.S.Means Company.
zation of such a database will be conducted in the future. Bubshait, A. A., and Cunningham, M. J. 共1998兲. “Comparison of delay
analysis methodologies.” J. Constr. Eng. Manage., 124共4兲, 315–322.
Finke, M. R. 共1998兲. “A better way to estimate and mitigate disruption.”
J. Constr. Eng. Manage., 124共6兲, 490–497.
Acknowledgment
Finke, M. R. 共1999兲. “Window analysis of compensable delays.” J. Con-
str. Eng. Manage., 125共2兲, 96–100.
The writers would like to thank the Korea Institute of Science and Kallo, G. G. 共1996兲. “Estimating loss of productivity claims.” J. Manage.
Technology Evaluation and Planning 共Grant No. M10104000274- Eng., 12共6兲, 13–15.
01J000012100兲 for their contributions to this research. Kartam, S. 共1999兲. “Generic methodology for analysing delay claims.” J.
Constr. Eng. Manage., 125共6兲, 409–419.
Kraiem, Z. M., and Diekmann, J. E. 共1987兲. “Concurrent delays in con-
Notation struction projects.” J. Constr. Eng. Manage., 113共4兲, 591–602.
Kululanga, G. G., Kuotcha, W., McCaffer, and Edum-Fotwe, F. 共2001兲.
The following symbols are used in this paper: “Construction contractors’ claim process framework.” J. Constr. Eng.
DC ⫽ duration within which planned work duration 共D P兲 Manage., 127共4兲, 309–314.
Reichard, D. D., and Norwood, C. L. 共2001兲. “Analyzing the cumulative
could be finished with daily average labors 共LU兲 and
impact of changes.” AACE International Transactions, AACE Inter-
unimpacted productivity 共PU兲; national.
DD ⫽ contractor’s difference between DC and D P; Richter, I. E.共1983兲. International construction claims: Avoiding & re-
Di ⫽ portion of ith duration due to ith factor; solving disputes, McGraw–Hill, New York.
DN ⫽ nonworked day; Rubin, R., Fairweather, V., Guy, S., and Maevis, A.共1992兲. Construction
DW ⫽ work delay; claims prevention and resolution, 2nd Ed., Van Nostrand-Reinhold,
DPL ⫽ lost productivity duration; New York.
Pi ⫽ portion of ith productivity due to ith factor; Shi, J. J., Cheung, S. O., and Arditi, D. 共2001兲. “Construction delay
PL ⫽ lost productivity; computation method.” J. Constr. Eng. Manage., 127共1兲, 60–65.

1154 / JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / NOVEMBER 2005