ICSDC 2011 335

On-Site Labor Productivity Measurement for Sustainable Construction

Ka Wah (Calvin) Chui1, Ph.D., P.E., M.ASCE, David Shields2, Ph.D., P.E., M.ASCE,
and Yong Bai3, Ph.D., P.E., F.ASCE
Downloaded from ascelibrary.org by UNIVERSITE LAVAL on 07/02/14. Copyright ASCE. For personal use only; all rights reserved.

ABSTRACT
One of the major barriers to advancing the competitiveness of sustainable
construction in the U.S. market is that traditional construction firms do not have
accurate data on labor productivity for sustainable construction projects. This
negatively impacts their ability to reliably estimate a construction projects schedule
and cost. Due to the nationwide diverseness and complexity of the construction
industry, labor productivity comparison is difficult. Even for conventional projects
minimal research has been performed on nationwide labor productivity measurement,
with most research being jobsite specific. This paper uses productivity data that has
been collected on jobsites in Kansas. Using the data, various statistical analysis
methods were applied to determine the productivity rate differences between two
projects and compare it to data from RS Means, which is an industry standard.
Statistical analysis results prove the possibility of using labor productivity to control
the labor cost of a green project. Furthermore, the results can be used by
construction firms entering any regional market as a basis for a comparative
reference. Through this advancement a firm may enhance their ability to perform
competitively in the sustainable construction market.

INTRODUCTION
The working-sampling method was employed to collect data for projects investigated
in this paper. The data were analyzed using statistical methods to determine if they
contain factor(s) that are related to labor productivity between jobsites. Specifically
investigated in this paper is labor-productivity data acquired for mechanical, electrical
and fire protection work. A carefully conceived plan was developed that ensured that
the exact data were collected that were required to perform necessary analyzes.
Workforce activities were characterized by three major categories: direct work (D)
using tools or effort at a designated work location to perform an assigned task that
makes a direct, productive contribution to completing the work scope; indirect work
(I) support activities that are not directly contributing to completing a job; and non-
working (N) all unexplained non-utilization or personal idle time.

Two ongoing construction projects were utilized to collect the necessary data for this
study. Both projects were located on the University of Kansas West Campus in

1
Assistant Professor, Construction Management Program, University of Nevada Las Vegas, 4505 S.
Maryland Parkway, Box 454005, Las Vegas, NV 89154-4005; PH (702) 895-1461; FAX (702) 895-
4966; e-mail: calvin.chui@unlv.edu
2
Director and Associate Professor, Construction Management Program, University of Nevada Las
Vegas, 4505 S. Maryland Parkway, Box 454005, Las Vegas, NV 89154-4005; PH (702) 895-1461;
FAX (702) 895-4966; e-mail: david.shields@unlv.edu
3
Associate Professor, Department of Civil, Environmental and Architectural Engineering, University
of Kansas, 1530 W. 15th Street, Lawrence, KS 66045-7526; PH (785) 864-2991; FAX (785) 864-5631;
e-mail: ybai@ku.edu

1
335
Copyright ASCE 2012 ICSDC 2011
ICSDC 2011
 336 ICSDC 2011

Lawrence, Kansas. The first project was a new building housing the School of
Pharmacy; the other was a new building housing the Bioscience and Technology
Downloaded from ascelibrary.org by UNIVERSITE LAVAL on 07/02/14. Copyright ASCE. For personal use only; all rights reserved.

Business Center. The new School of Pharmacy building was a $45.5 million project
that was started in late May 2009, approximately 60 percent was complete when data
collection was initiated, and it was completed in late July 2010 (see Figure 1). The
project involved the construction of a 10,220 square meter, four-story, steel-framed
building plus a level of basement and exterior glass-and-brick-faced structure. The
structure had a mainly curtain wall faade. The second construction project, the
Bioscience and Technology Business Center, was a $7.25 million project that was
started in early October 2009. Approximately 20 percent was completed when data
collection was initiated, and it was completed in early July 2010 (see Figure 2). The
1,858 square meter, two-story, steel-framed building facility had mainly glass-and-
brick-faced exterior walls. The data collected at the two jobsites, for this study,
includes data from the heating, ventilating, and air conditioning (HVAC) system, fire
protection sprinkler system and masonry work.

Figure 1. School of Pharmacy Building (May 2010)

Figure 2. Bioscience & Technology Business Center (May 2010)

DEFINITIONS OF CONSTRUCTION PRODUCTIVITY

A frequent construction-industry definition for productivity is: to measure the
effectiveness with which management skills, workers, materials, equipment, tools and
working space are employed at, or in support of, work-face activities, to produce a
finished building, plant, structure or other fixed facilities at the lowest feasible cost
(Oglesby et al., 1989). Dozzi and AbouRizk (1993) defined labor productivity as

336
2
Copyright ASCE 2012 ICSDC 2011
ICSDC 2011
 ICSDC 2011 337

measured at an activity level, and because construction activities are normally labor
intensive, productivity at the activity level is frequently referred to as labor
Downloaded from ascelibrary.org by UNIVERSITE LAVAL on 07/02/14. Copyright ASCE. For personal use only; all rights reserved.

productivity, which measures the input as labor hours and the output as installed
quantities. Following this definition labor productivity is expressed as the ratio of the
quantity of input to the quantity of output.

Labor Hours
Labor P roductivit y =
Completed Work (Unit)

When labor productivity is defined in a detailed manner, as it is later in this paper, it

is measured in actual work hours per installed quantity. As the labor-productivity
value decreases, performance improvement is being achieved. Productivity also has
another definition, including performance factors production rate, and unit person
hour rate. Generally in construction, productivity is stated as an in-place value
divided by inputs, such as work hours. Although most project owners and contractors
adhere to the definition of productivity as dollars of output per dollars of input or an
increase in sales, this definition is not widely accepted (Adrian, 2004). Therefore,
thereby substituting dollars for person-hours of input, labor productivity is the ratio of
physical output per unit of work hour requirements as expressed in the following
equation.

Physical Output (Units)

Labor Productivi ty =
Labor Hours

RESEARCH ON PRODUCTIVITY COMPETITIVENESS

Construction productivity is a way of measuring how is produced by a given amount
of resources or how much resource is needed when producing a given number/volume
of a product. The reason for measuring productivity is to understand the production
processes and learn about the capacity of machinery and workers (Ingvaldsen et al.,
2004). Productivity measurements quantify how efficient resources are used and to
provide a numerical indicator of the performance ability of companies/project
owners/contractors in a competitive market.

To further the construction industry knowledge base, the primary goal of the research
reported in this paper was to conduct an accurate measurement of on-site construction
productivity for comparison of labor productivity at the task level. To achieve this
goal on-site data collection was conducted; data analysis performed, comparisons
made based upon the effectiveness of building construction based on the productivity
results compared with industry-standard estimating data, and finally provide a
summarization of research results and guidelines for future development of on-site
construction productivity measurement. It is widely accepted that productivity
measurement plays an important role in the management of construction. Productivity
measurement provides the necessary data to analyze factors for project owners,
constructors, and management professionals to control construction progress, estimate
the cost of future construction projects, and determine its competiveness in the global
market. In achieving these objectives, the researcher hopes that the results will

337
3
Copyright ASCE 2012 ICSDC 2011
ICSDC 2011
 338 ICSDC 2011

contribute to enhancing a firms ability to perform competitively in the sustainable

construction market.
Downloaded from ascelibrary.org by UNIVERSITE LAVAL on 07/02/14. Copyright ASCE. For personal use only; all rights reserved.

DATA COLLECTION PROCEDURE

Data were collected through observations during daily site visits to the two previously
described building construction projects. The data collection period was from
February 2010 to April 2010. The data collection procedures were established in
collaboration with the construction firms personnel and staff members of Design &
Construction Management Office in the University of Kansas. The selection of
projects jobsites and work items for observation were investigated following the
procedure outlined in Figure 3. For each candidate project proposed for this study,
each project was prescreened and specifically excluded if the project was in the early
phase of the work or the startup/commissioning phase of the project since these stages
of construction would affect labor productivity in ways that would introduce
unwanted variability into the data and would be difficult to account for. The
following items were also included for consideration: (1) Data collection
methodology; (2) Safety protocol for site visits; (3) Research assistants training and
availability for on-site observations; (4) Selection of work items to be observed; and
(5) Site visit schedule for each project.

Identify Project and

Owner

Project Meets
Prescreen Phase Selection No
Criteria

Yes

Obtain
Owner Approval for No
Research Project Characteristics
- Safety Protocol
- Project Diversity (Type, etc.)
Yes - Accessible Project Documents
- Site Personnel to be Contacted
Conduct - Activity Status
Initial Site Visit

Identify Collectable
Work Items

Daily Site Data

Collection

Daily Organization
of Data for Analysis

Figure 3. Process to conduct labor productivity data

338
4
Copyright ASCE 2012 ICSDC 2011
ICSDC 2011
 ICSDC 2011 339

Data collection focused on a task-level productivity measurement per 10-minute work

cycle, specifically, the on-site measurement was accomplished by randomly selecting
Downloaded from ascelibrary.org by UNIVERSITE LAVAL on 07/02/14. Copyright ASCE. For personal use only; all rights reserved.

a worker in a single shift and the documentation of the factors that may have affected
work. It was important to determine the process of a work activity from the beginning
until it was finished in a work cycle because any large variation in a job task would
have complicated the data collection procedures. Since the daily work process and the
work environment were equally important, the job task had to be accessible for
physical and visual inspection. The daily measurement during the work shift involved
counting work hours, quantities and quality of work items installed and then
documenting other factors regarding the job site. Discussions with the workers,
especially the crew/foremen, were held frequently to obtain information regarding the
progress and problems encountered by the workers. Continuous variables were
measured daily, which include crew size, workers age and experience, temperature,
weather condition, interference, and so forth. It required the observers to be familiar
with the site conditions and have a good working relationship with the crew, foremen
and superintendents. Work sampling coding was divided into three categories which
contained various subcategories. The subcategories and their definitions are provided
in Table 1.

Data collection was based on visual observations, physical measurements, reviews of

construction documents, and brief discussions with the workers, crew or foremen.
Data collection forms were provided to the observers for recording the data and other
factors that may have affected the productivity. A data collection form was used to
specify the scope of each work item for which data were collected and to note
different factors that may influence the production rate of each work item at task-
level as shown in Table 2. It provided guidance to ensure consistent observations and
data collection. Work elements included in the work sampling were those that most
directly represent actual production of the work item and are the primary concerns in
estimating construction time. To accommodate variability in the work sampling and
task-level factors among work items, each data collection form is unique for a given
work item.

Table 1. Work sampling coding used for on-site data collection.

Direct Work (D) Indirect Work (I) Non-Working (N)
D1: Measure space for I1: Read blueprint drawing N1: Chat with others
exact position before I2: Prepare space for direct work (observing, initial N2: Personal reasons
installation measuring, cleaning, clearing, marking, cutting (phone calls,
D2: Prepare materials for opening, etc.) smoking, restroom,
the activity (material I3: Walk back/relocate with tools/materials etc.)
measuring, cutting, (equipment, hand/power tools, accessories, etc.) N3: Break time
lifting, taping, I4: Walk back/around empty handed (searching for N4: Early Leave for
modifying, etc.) tools/materials/accessories, etc.) Lunch
D3: Hands-on activity I5: Discuss with foremen/co-worker for direct work
(installing, sealing, I6: Receive tools/materials/assistance from other
finishing, etc.) workers
D4: Check and adjust I7: Assist co-worker
position/alignment of I8: Re-adjust previous installation
new installation I9: Get electrical power for tools/equipment
I10: Seal outlet/opening of installment for temporary
protection

339
5
Copyright ASCE 2012 ICSDC 2011
ICSDC 2011
 340 ICSDC 2011

Table 2. Work sampling item: Included vs. Not Included.

Scope
Work Item Included Not Included
Downloaded from ascelibrary.org by UNIVERSITE LAVAL on 07/02/14. Copyright ASCE. For personal use only; all rights reserved.

Fire Protection System Installation of pipes, supports, moving
Fire Protection Sprinkler equipment/material, installation of
System valves/heads, on-site
preparation/cleaning before work, and
breaks
Material fabrication and
transportation/cleaning site after
work/inspection & testing

Building Mechanical Installation of ductwork/equipment, Material fabrication and

System Heating, supports, moving equipment/material, transportation/cleaning site after
Ventilating, and Air on-site preparation/cleaning before work/inspection & testing
Conditioning (HVAC) work, and breaks
System

Masonry Brickwork Placement of brick works, on-site Cleaning site after work/
preparation/cleaning before work, and transportation of bricks from jobsite
breaks yard to work location/material
handling/mixing mortar

DATA COLLECTION AND ANALYSIS RESULTS

Statistical methods were applied to identify the causes of differences, if any, after
comparing the data collected and to illustrate the methodology of data presentation.
The factors were identified by visually inspecting scatter plots and distribution plots
and by using descriptive statistics. An analysis of variance (ANOVA), frequency
analysis, correlation, Chi-Square test, and Kruskal-Wallis test were then employed to
analysis the data by using the Statistical Package for Social Scientists (SPSS).

Descriptive statistics are commonly used to summarize a data set in a simple and
understandable way, rather than being used to support inferential statements about the
measurements or values that the data are thought to represent. The data gathered were
analyzed to present the number of work items in each category and its corresponding
work cycle time. The data sets contain data covering a total of 239.3 work hours from
a total of 1,436 observations (10-minute work cycles). Comparison of work hours for
all three work categories is shown in Table 3.

Table 3. Data set summary statistics for categories.

Direct Indirect Non- Total
No. of
Category Work Work Working Work
Observations
Hours Hours Hours Hours
Fire Protection Sprinkler System 442 27.11 37.41 9.15 73.67
HVAC System 407 32.79 26.35 8.69 67.83
Brickwork 587 66.23 22.21 9.40 97.83

340
6
Copyright ASCE 2012 ICSDC 2011
ICSDC 2011
 ICSDC 2011 341

100%
Downloaded from ascelibrary.org by UNIVERSITE LAVAL on 07/02/14. Copyright ASCE. For personal use only; all rights reserved.

80%

60% Non-Working
Indirect Work
Direct Work
40%

20%

0%
Fire Protection HVAC Brickwork
Sprinkler System System

Figure 4. Work category proportions by work category

SUMMARY AND CONCLUSIONS

In this study, a total of 239.3 work hours of on-site data measurement were
investigated to determine the work proportions of each work category. All data were
analyzed so that direct work hours, indirect work hours, and non-working hours for
each operation could be determined. The work category which had the highest direct
work hours compared to the other two work categories was brickwork which showed
fewer indirect work hours and non-working hours. Comparing the other two work
categories, which include fire protection sprinkler system and HVAC system, it
shows a slightly different output with the HVAC system being higher than the fire
protection sprinkler system. From the results of analyzing the sub-categories of each
operation more differences could be identify between the three work categories. Data
collected from the jobsites indicate that between 37 and 67 percent is for direct work
hours. In addition, the time proportions are illustrated which includes the direct work
hours (37-67 percent), indirect work hours (23-50 percent) and non-working (10-13
percent) hours.

As shown in Table 4, the data comparison of labor productivity shows that the
workers had the labor productive rate of an average 5.73 meter per hour for pipe
installation in the fire protection sprinkler system work category. As for HVAC
system, the workers had the labor productive rate of an average 2.78 meter per hour
for duct installation. Last but not least, for brickwork, the workers had lower labor
productive rate of 61.08 bricks per hour versus the RS Means standard rate of an
average 64.69 bricks per hour in Kansas.

Table 4. Comparison of labor productivity results.

U.S. RS Means
Fire Protection Sprinkler System 5.73 m/hr N/A
HVAC System 2.78 m/hr N/A
Brickwork 61.08 bricks/hr 64.69 bricks/hr
Brickwork (Based on Face Area) 0.677 m2/hr 0.717 m2/hr

341
7
Copyright ASCE 2012 ICSDC 2011
ICSDC 2011
 342 ICSDC 2011

The results of this research can lead to a comparative review of the fraction of
working or nonworking time in each work category and the reasons for it can lead to
Downloaded from ascelibrary.org by UNIVERSITE LAVAL on 07/02/14. Copyright ASCE. For personal use only; all rights reserved.

actions by both labor and management to improve productivity by reducing the time
spent on indirect and non-working activities. Furthermore, it could also help to enable
cost management to affect productivity improvement on sustainable construction
projects. This research also provided an example for systematic statistical observation
on a project which can help monitor the work activity to obtain an overall picture of
the utilization of the workforce. As part of its discipline, inferences can be made
regarding constraints to the flow of work and resulting inefficiencies in the process.

The method presented in this paper can be used to measure labor productivity on
sustainable construction projects, i.e. Leadership in Energy and Environmental
Design (LEED), Green Globes, BRE Environmental Assessment Method
(BREEAM), etc. Sustainable construction has a number of definitions, none of which
would preclude the use of the methodology presented in this paper. Applying this
methodology in the same manner with the same approach would yield a comparison
of conventional construction versus construction based upon sustainable design
guidance. Sustainable construction is generally considered to be more of a life-cycle
approach that is environmentally and resource conscientious. We are concerned with
labor productivity comparison between the construction phases and conventional- and
sustainable-based design of a building. Labor productivity measurements offer a more
penetrating investigation into a possible area of fundamental difference between the
two approaches. Ultimately, improved labor productivity can reduce building cost.
During most of the past century much emphasis has been placed on improved labor
productivity employing near-centuries old techniques. Henry Gantt, of Gantt chart
fame, perhaps was one of the first to scientifically investigate masonry labor
productivity. Recent movements such as lean construction have made great strides in
improving labor productivity. Due to federal government and state/local governments
enacting sustainable construction into law for new capital facility construction this is
more than a passing fad. Therefore, investigations into labor productivity need to be
launched to determine the impact on labor productivity of sustainable design
practices.

As this study identified the work category, brickwork, with the least indirect and non-
working hours was an area that has had probably the fewest technological advances in
several centuries. Areas such as HVAC construction will be impacted with
complexities and unknown means and methods based upon technological advances
incorporated by manufacturers and designers in search of energy efficient designs
which support sustainability by reducing energy usage, use renewable energy sources
to reduce carbon footprint, etc. The likelihood of reduced labor productivity is real in
trades involved in HVAC.

REFERENCES
Adrian, J. J. (2004). Construction Productivity: Measurement and Improvement.
Stipes Publishing L.L.C., Champaign, IL.

342
8
Copyright ASCE 2012 ICSDC 2011
ICSDC 2011
 ICSDC 2011 343

Associated General Constractors of America (AGC). (2010). The Economic Impact

of Construction in the United States. <http://www.agc.org/galleries/econ/
Downloaded from ascelibrary.org by UNIVERSITE LAVAL on 07/02/14. Copyright ASCE. For personal use only; all rights reserved.

National%20Fact%20Sheet.pdf> Updated in 2010, Accessed on 12/2010.

Chapman, R. E., and Butry, D. T. (2009). Measuring and Improving the Productivity
of the U.S. Construction Industry: Issues, Challenges, and Opportunities, Building
and Fire Research Laboratory, Nation Institute of Standards and Technology.

Dozzi, S. P., and AbouRizk, S. M. (1993). Productivity in Construction. Institute for

Research in Construction, National Research Council, Ottawa, ON, Canada.

Ingvaldsen, T., Lakka, A., Nielsen, A., Bertelsen, N. H., and Jonsson, B. (2004).
Productivity Studies in Nordic Building- and Construction Industry. Norwegian
Building Research Institute, Oslo, Norwegian.

Picard, H. E. (2004). Direct Labor Productivity Measurement As Applied in

Construction and Major Maintenance Projects. TCM Framework: 9.2 Progress and
Performance Measurement. AACE International Recommended Practice No. 22R-01.

Oglesby, C. H., Parker, H. W., and Howel, G. A. (1989). Productivity Improvement in

Construction, McGraw Hill, New York, U.S.A.

343
9
Copyright ASCE 2012 ICSDC 2011
ICSDC 2011