Kaplan05analysis

Proceedings Construction Research Congress 2005, ASCE, San Diego, California, 5-7 April 2005

ANALYSIS OF VARIABILITY IN PRECASTING AND
INSTALLATION OF PILE FOUNDATIONS
Hasan Kaplan1, Andries Elburg2, and Iris D. Tommelein3
ABSTRACT
This paper presents a study on the use of pre-stressed precast foundation piles using data
collected on a building project that comprised the installation of more than 340 piles. The
objective is to assess the presence of product and process variability in demand and
supply, and the impact it has on precast pile delivery. Sources of variability in precasting,
transportation, and installation are described. As a step towards understanding the
complexity of this production system and to show how production planning decisions
affect execution, this paper elaborates on two deterministic scenarios and compares those
with actual data from the project. Performance of the system is analyzed on the basis of
inventory vs. work completed. From the perspective of ‘lean’ thinking the paper then
identifies sources of waste and suggests means to eliminate them. It also presents
questions for follow-on research.
KEY WORDS
Pre-stressed precast piles, production management, deep foundation construction,
variability, standardization, supply chain, lean construction
INTRODUCTION
Placing a foundation is one of the first activities on a construction site: a foundation
literally forms the basis of the rest of the construction work. One way of building a
foundation is to design and precast concrete piles, transport them, and drive them into the
ground. A case study is presented here that pertains to an 8-story reinforced concrete
residential building in San Francisco, California, with a foundation made of piles that
were precast at a plant in Antioch, California. The authors interviewed the industry
practitioners involved in the production and installation of these precast piles. We
identified various occurrences of variability and uncertainty in this production system.
Because of variability and uncertainty, generally speaking, lead times tend to increase
and schedules risk being jeopardized. We present a deterministic model and analysis and
then suggest potential system improvements based on ‘lean’ thinking (Ballard et al. 2002,
Koskela et al. 2002). Follow-on research will include stochastic modeling and analysis.

1

2

3

Student, Mechanical Engineering, University of Twente, the Netherlands and visiting
student researcher, University of California at Berkeley, h.kaplan@student.utwente.nl
Student, Industrial Management, University of Twente, the Netherlands and visiting
student researcher at Univ. of California at Berkeley, a.elburg@student.utwente.nl
Professor, Engineering and Project Management Program, Civil and Environmental
Engineering Department, 215-A McLaughlin Hall, University of California at
Berkeley, CA 94720-1712, Tommelein@ce.berkeley.edu
1


RELATED WORK
Foundation construction is notoriously challenging because of uncertain subsurface
conditions that lead to a significant amount of variability in the delivery system. Lean
production regards undesired variability as a major culprit for waste (Hopp and Spearman
2000). In order to see the causes of such waste and identify means to avoid them, one
needs to reveal and articulate the intricacies of the production system, then manage those,
because ‘the devil is in the details.’ Detailed descriptions of observed practices through
case studies, analysis, and testing against theory, are therefore in order (e.g., the case
study presented here). In contrast, many descriptions in the construction literature remain
high-level in order to maintain a sense of generality (e.g., Peurifoy et al. 1996).
Presumably generality cannot be obtained at a more detailed level, but this could be the
consequence of lack of theoretical understanding.
The study of a precast pile production system, as described here, relates to lean
construction studies that illustrate how competitive advantage in the construction industry
may be achieved by identifying opportunities for lead-time reduction in engineered-toorder products (e.g., Ballard 2001, Ballard et al. 2003, Arbulu and Tommelein 2002,
Elfving 2003, Tommelein and Ballard 2005). Lean construction tools and techniques,
such as understanding interacting sub-cycles (Howell et al. 1993), designing construction
operations using first-run studies (Howell and Ballard 1999), standardizing products,
using tight real-time process feedback loops (Tommelein 1998), and strategic positioning
of buffers and batch sizing (Alves and Tommelein 2004, Walsh et al. 2004) apply when
looking for system improvements in the delivery ofprecast piles.
DESIGN OF FOUNDATION PILES
The focus here is on pre-stressed precast piles. These are formed by tensioning highstrength steel cable before casting the concrete. After the concrete has developed
adequate strength the cables are cut, so they will apply a compressive stress on the pile.
Typically, ‘a group of piles is used to transmit the load of the superstructure to the soil
mass.’ (Brown 2001). ‘The individual piles are spaced and connected to the pile cap or tie
beams and trimmed in order to connect the pile to the structure at cut-off level, and
depending on the type of structure and eccentricity of the load, they can be arranged in
different patterns’ (Napier 2004).
On the project in this study, the geotechnical engineering (GE) firm recommended to
the owner that precast pre-stressed precast piles be used (Treadwell&Rollo 2002). At a
later time, upon the request from the general contractor (GC), the GE also studied the use
of steel H-piles but these proved to be less economical at the time of decision making.
The characteristics of a pile depend on the loads the pile may be subjected to and the
loading capacity it can develop. The pile design process has a reciprocal dependence and
is thus iterative. The SE determines the number of piles based on column loads
(determined by the SE) and pile capacity (determined by the GE). The GE decides on the
piles’ cross sections (typically in collaboration with the SE) and their lengths.
This project included placing a foundation of 341 piles. On average, 8 +/- 3 piles per
day were driven into the ground, a number that increased to 12 or 13 piles per day or
even more when conditions were favorable. A unique challenge of the site was that the
bedrock, in which the foundation piles had to be grounded, was not level. The GE drilled
five test borings for this specific project, reviewed drilled borings (and possibly CPTs)

2


from a previous investigation as well as data from older investigations by others, and also
relied on their experience with projects on adjacent sites to design the foundation system.
Early on during construction, they conducted a ‘first-run study’ to determine the location
of the bedrock more precisely: twenty indicator piles were driven into the ground. These
confirmed that the depth of this bedrock varied anywhere from -25 m down to -55 m (-83
ft down to -180 ft): a challenge for a rectangular site lot with dimensions 55 m by 73 m
(180 ft by 240 ft). While these indicator piles were made specifically to obtain test data,
they also served in the foundation system being built.
PILE PRODUCTION
The pile production process consists of the following steps:
• Adjusting formwork: The precaster in this study had casting ‘yards’ 260 m (850 ft)
long, with steel forms and blocks to be set to the appropriate pile width and length.
• Making caging: After the distances of the blocks are determined, reinforcement is
installed. First rebar coil is inserted in the form, cables are pulledthrough them,
longitudinal rebar is added, and then the cables can be stretched. Subsequently the
coil is stretched and the longitudinal rebar is manually lifted into its final position.
The rebar is precut and supplied in the right sizes, but tying it is time-consuming.
• Mixing concrete: The mixing process is completely computerized. It only takes on
the order of two minutes to prepare a batch ready to pour.
• Pouring concrete: Pouring must happen within 30 minutes or so of mixing, but
usually it is completed in as little as 10 minutes. A truck with a screw auger transports
the ready-to-pour concrete. After pouring the concrete in the formwork it has to be
vibrated and smoothened.
• Hardening concrete: The concrete hardens overnight thanks to the use of steam
curing (steam blankets).
• Removing concrete from formwork: A crane lifts the pile out of the form and puts
it in the precaster’s inventory. It remains there for at least 14 days or until design
strength is reached (typically 41 MPa (6,000 psi)) to allow for transport and driving.
PILE INSTALLATION
The pile driving process consists of the following steps:
• Pre-drilling: A hole is drilled about 6 to 9 m (20 to 30 ft) deep, typically the depth of
the fill, to pre-position the pile and remove possible obstruction/debris. This step
involves an auger operator and a worker who guides the process from the ground.
• Pile handling: taking the pile out of on-site inventory and lifting it up causes the
most stress on the element, when it is lifted from the horizontal to the vertical position
by the pile driver. This step involves the piledriver operator, two people to attach
cables to the pile, and someone who guides the process.
• Connecting pile to hammer: Rebar that sticks out at the top of the column has to fit
into holes in the head of the hammer. When fitting the rebar, one person operates the
pile driver. The operator gets guidance from someone on the ground who is better
able to see the hammer. Another worker tries to steer and turn the column in the right
direction from the ground, while another person supervises the process. This step

3


•

•

requires skill from the pile driver operator. If it fails, a worker has to climb up and fit
the rebar manually, which may cause delays in the operation on the order of 1/2 hour.
Placing pile in pre-drilled hole: The pile driver drops the pile into the pre-drilled
hole in preparation for driving. When placing the pile in the hole it sometimes sinks
on its own as much as 15 m (50 ft) or so through the layer of Bay Mud. This step
involves four people: a pile driver operator, a person who turns the pile to the right
angle, a person to signal to the operator, and a supervisor.
Driving pile in ground: The pileis hammered into the ground until it progresses less
than 60 blows/30 cm (60 blows per ft). This refusal criterion is to make sure the pile
is in sufficiently hard rock to support the load (not necessarily at the rock surface).
The supervising GE counts the blows and decides when the final depth is reached.
This step involves the pile driver, the GE, and a worker who holds the rope to shut off
the fuel supply. It may vary in time, depending on subsurface conditions.

PLAYERS IN THE FIELD AND THEIR INTERACTION
Thee contractual relationships for pile delivery on this project—which are fairly common
for the industry—involved many different supply-chain participants. The owner retained
the GE early on in this project and under a separate contract to get soils analysis data. The
SE determined the loads and stresses that act on the piles, and accordingly designed the
foundation system. That design was passed on to the architect, who in turn passed it on to
the GC. The GC then requested design-build proposals from several pile driving
companies, who in turn requested quotes from precasters, and selected one. The precaster
developed a design solution based on the data from the SE and GE. This design had to be
approved by the SE and by the GE for conformance with design intent so it was passed
back along contractual relationship links. The GE determined the pile tip elevations,
considering variations in the depth and hardness of the bedrock across the site, and ended
up specifying a 10 m (15 ft) cut-off. Based on this data, the pile driving contractor
determined the pile lengths. Clearly, a lot of communication is necessary to ensure a
comprehensive design- and construction process.
PRODUCT- AND PROCESS VARIABILITY
Different kinds of variability influence the piledelivery process, for example:
1. The design process is fragmented. The SE designs the piles, but the GE provides
substantial input, and a designer from the precast plant develops the design and works
out the details. Both the GE and the SE review the pile shop drawings and may
provide feedback to the precaster.
2. The production process appeared to have a low degree of variability in time. It was a
fairly streamlined and reliable process, based on the precaster’s standard work sheets
that spelled out the detailed process steps.
The normal curing time for concrete is 14 days but by adding an accelerator to the
concrete mixture the precaster can decrease this time by a few days. The owner, GC,
pile driver or precaster decide on the use of (and pay for) this additive in order to
accelerate the schedule.
3. The pile driving process was highly variable. Circumstances on site are difficult to
assess and control (subsurface conditions, access restrictions and obstructions,
reliability of equipment, weather, labor skill and availability). Data collected on a

4


productive day, during which 11 piles were installed, showed the duration ranging
from 21 min for one pile to 84 min for another pile, starting with removing the pile
from on-site storage until driving it completely into the ground.
4. Variability in demand posed the biggest challenge. The GC manages the entire
construction project and decides which schedule best suits the owner’s needs, but
since pile driving is such a crucial activity early on in the project, the pile driving
subcontractor essentially sets and controls this early part of the schedule. This
notwithstanding, the project schedule changes and changes must be dealt with as they
arize. For example, when a pile breaks it must be replaced. As a contingency, the pile
driver or the precaster may have piles (from other projects or made extra) that can be
used for this purpose. Some design variance is granted in this circumstance in order to
accommodate the immediate needs; otherwise, the precaster would have to produce a
replacement for the broken, engineered-to-orderpile.
If not accounted for during planning, a different pile sequence can jeopardize the
installation schedule of the pile driving company and also the production schedule of
the precaster. The lead time for making a pile is about 3 weeks (incl ding 2 weeks for
u
curing), so if there is no appropriate pile in inventory, a change in project schedule
could cause construction to come to a halt. To avoid this, the piledriving
subcontractor may decide to drive a “green” pile (<14 day cure) at their risk.
REDUCING VARIABILITY
Supplier Involvement in Design: On this project, some iteration was needed in the
precaster’s design process, involving both the SE and GE. Koskela et al. (1997) noted
that roughly 50% of the causes for uncompleted design tasks are related to variability in
the design process. Reducing this variability can reduce the duration and improve the
quality of the design process. “An alternative approach is to involve specialty contractors
early on in the design process to take advantage of their ability to develop creative
solutions and their knowledge of space consideration for construction processes,
fabrication, and construction capabilities and supplier lead times and reliability” (Gil et
al. 2001). “Suppliers are no longer viewed as vendors of a commodity, but rather are
brought into the picture as true partners in the design and production of the end product”
(Mikulina 1998). The GE pointed out (personal communication) that coordination
meetings involving themselves, the SE, and the precaster are seldom called, yet they can
be effective in communicating design intent.
Product and Process Standardization: Standardization can reduce the duration of
design because it reduces the amount of communication between parties. It also can
simplify the production process. On this project, all piles were 35.6 cm by 35.6 cm (14”
by 14”) square so formwork did not have to be readjusted. All compression piles were
standardized to a large extent, and so were the tension piles, even though the piles’ length
and rebar configuration varied. Could standardization have been taken further?
Standardized pile designs work, e.g., for CalTrans, because they build relatively similar
structures with similar loads (e.g., highway overpasses) and their engineers use the same
design manual. Engineers for other types of facilities often prefer to create value through
custom design because it affords them the opportunity to fine tune their design based on
specific project requirements (GE personal communication).

5


Transportation: The pile driving company is responsible for transporting piles from the
precaster to the site. This requires planning because piles are heavy and those longer than
80 feet require a transportation permit from the city of San Francisco. The City restricts
when transport can take place, taking into account rush hour and other traffic conditions.
Another limiting factor is the availability of trailer trucks and drivers qualified to handle
the extremely long loads.
Space on site for storing and manipulating piles was limited and further diminished as
foundation work progressed, so keeping a big inventory was not possible. These factors
made reliable transport of piles challenging. Loading and off-loading of the piles was a
relatively straightforward, more repetitive process.
Installation: The pile driving company schedules the installation work. They set the
sequencing of piles based on site access constraints and in consultation with the GC. The
company estimates the duration needed to drive a pile based on the soils report,
equipment selection, site conditions, and experience. It assumes the diesel hammer makes
a certain number of blows per minute, and it then relates blows to soils data (e.g., SPT).
The company also includes an efficiency factor to account for site logistics and other
factors, and looks at previous jobs conducted in the area.
The pile driving schedule is coordinated with the precaster’s schedule. To allow for
variability in site work, the precaster tries to get ahead of installation work by producing
piles, curing them, and keeping piles in inventory. This makes it feasible to install piles
more quickly than scheduled and it provides the flexibility to swap the order in which
piles get installed should project changes require it.
Standardization of piles makes piles more interchangeable. If one pile would break a
perfect substitute replacement would be available.
Site space availability is even more important with regards to installation than it is for
storage. Limited space hampers equipment movement and makes it more likely for pile
driving to interfere with other subcontractors’ work.
Demand: Demand variability in pile driving is caused in part by the fact that the
sequence in which work gets done affects how much work is to be done. Displacement
piles, such as the piles in this study, push soil out while they are being driven into the
ground. This makes it more difficult to subsequently drive other piles in the vicinity.
Demand variability may also be caused by changes in the project schedule, such as
those stemming from the manifestation of uncertainty in the installation process. The
sequencing of pile driving was changed on this project, due to heaving of a sidewalk
pavement observed at the edge of the site. The GE stopped site work to investigate the
situation. They speculated that piles might be pushing soil towards the China Basin
Channel. As a countermeasure, they ordered that holes for the piles be pre-drilled deeper
(down to 10 m (30 ft)) and larger so more soil would be dug up. This changed the
subcontractor’s work method, as larger holes could not be left unattended (smaller holes
for a day’s worth of work could be drilled all at once). Pre-drilling now had to be
followed immediately by pile installation so the auger and its operator had to remain on
stand-by. The GE also ordered relief holes drilled near the pile cap at the side of the
Channel so displaced soil could be pushed into them, and to be filled after completion of
the pile driving work.

6


ALTERNATIVE MODEL SCENARIOS AND CALCULATION OF INVENTORY
As a step towards understanding the complexity of this production system, we introduce
simplifications, develop two model scenarios, analyze their inputs and outputs, and then
compare them against actual site data. Our aim is to explore how production planning
decisions affect process durations, as well as inventory at the precaster’s and on site.
Assume: (1) all processes progress at a steady rate (modeling of uncertainty and
variability is left for follow-on research); (2) piles take two weeks to cure (an equivalent
of 10 working days) in order to be strong enough for transportation; (3) 8 piles/day are
driven into the ground; (4) 8 piles/day are transported to the site to meet this installation
rate; and (5) space on site is limited and allows for 8 piles maximum. Tables 1 and 2
show, for each scenario, (1) which day it is, (2) the number of piles cast on that day, (3)
the precaster’s cumulative inventory at the end of the day, (4) the number of piles
transported to site on that day, (5) the cumulative number of piles stored on site, and (6)
the cumulative number of piles driven into the ground.

0
8
16-336
344
352
357

16
16
16
16
16
5

8
8
8
8
8
8
8
8
8
8
8
8
5

8
8
8
8
8
8
8
8
8
8
8
8
5
5

Installed

16
16
32
32
224
224
232
224
232
224
232
224
221
213
205
197-21
13
5

Stored On
Site

16

Transported

Day

Inventoried

8
8
8
8
5

1
2
3
4
5-27
28
29
30
31-41
42
43
44
45
46
47
48-70
71
72
73
74

Produced

8
8
8
8
5

Table 3: Scenario 2
(average daily inventory = 137 ± 77 piles)

Installed

51
102
153
204
255
306
357
357
357
357
349
341
333-13
5

Stored On
Site

51
51
51
51
51
51
51

Transported

Inventoried

1
2
3
4
5
6
7
8
9
10
11
12
13-53
54
55
56

Produced

Day

Table 2: Scenario 1
(avg. daily inventory = 184 ± 112 piles)

0
8
16-96
104
112
120
128
136
144
152-328
336
344
352
357

In scenario 1 (Table 1), piles are cast in the order needed using the maximum daily
production capacity (51 piles/day assumed). The precaster completes production three
days before transportation to site starts, at the rate piles are needed (8 piles/day).
Numbers as shown do not distinguish indicator piles from other piles, and the total
number accounts for each section of pile (very long piles were made in two sections).
In scenario 2 (Table 2), which more closely describes the precaster’s practice, piles
are made on alternate days at a rate of 2 days worth of installation per day of production
(16 piles/2 days). Production starts ahead of construction so as to have 1 month worth of
inventory at all times.

7


Actual data for production and placing was obtained from records provided by the
precaster and by the GE (not show due to space limitations). Note that the numbers vary
slightly because some piles were cast in two sections, but they count as one installed.
Figure 1 depicts these three sets of data. Comparison of the scenarios yields the
following observations. The total duration for scenario 1 is 56 working days, but
scenario 2 takes nearly four more weeks (74 working days). Transportation of the piles
starts later, because the precast plant wants a 4-week lead on the pile-driving schedule.
The maximum as well as the daily average amount of inventory differs in the scenarios.
Scenario 1 presents one extreme. Scenario 2 shows the benefits of synchronizing pile
production with pile driving. It is more of a ‘just-in-time’ system, though further
improvements are still possible.
Figure 2 shows lead times for pile production, assuming all materials are on hand
when production starts. The arrows above the time line reflect a scenario that has a 7
week lookahead window. The arrows under the time line reflect a ‘just-in-time’ scenario
that has a 3 week lookahead window.
Number of Piles

400
350
300
250

S1 Inventory

S1 Installed
Actual
Installed

Actual
Inventory

S2 Installed

200
150

S2 Inventory
100
50
0
1

11

21

31

41

51

Time 61

[days]

Figure 1: Piles in Inventory and Total Installed
for Two Scenarios vs. Actual

Figure 2: Lead Times for Production
Scheduling vs. Installation Scheduling

The advantages of a ‘just-in-time’ system are several, especially for projects with
changing demands:
• A smaller inventory of finished products means a smaller storage and handling cost.
• Less space needed means a lower capital investment (or excess space could be
rented), lower taxes (e.g., state and property taxes), and lower insurance rates to be
paid on inventories (even though the precaster gets paid for piles ready for shipment).
It could also make it easier to find a good location for a new precasting plant.
• Smaller risk for obsolescence, which stems from the risk that a product will lose some
of its value because of changes in the project or market. If design changes are likely
in the course of a project, then it is better to hold off on making products, because
once made, products are much more costly to change if they can be changed at all.
• Reduced lead time from order-to-receipt and increased responsiveness of the
production system to changing site needs, so less inventory needs to be maintained
for the sake of flexibility.
QUESTIONS FOR FOLLOW-ON RESEARCH
Questions thatwill guide follow -on research include:

8


1. What is the impact of variability on execution? This paper presented a deterministic
model and assumes there are no ‘matching’ problems (Tommelein 1998). The model
must be developed further to reflect product variety and other variability: How does
the length of each pile affect how piles are sequenced for fabrication? How does the
precaster define batches of ‘like’ piles? How much opportunity does the precaster
have to make piles more standardized (e.g., in terms of rebar lengths and sizes)? If
piles are too long to be transported in a single unit and thus have to be spliced, how
does that affect transportation and installation? It must also be developed further to
reflect the stochastic nature of modelparameters (e.g., durations).
2. What are the costs and benefits for driving indicator piles? How does one go about
choosing where they should be driven and how many of them?
3. What and how much waste is there in current practice? For example, based on the
significant amount of uncertainty with regards to the depth of bedrock and
considering the cost for making a pile longer should that prove to be necessary after
they have been driven into the ground, the GE justifiably specified relatively large
cut-off lengths. Notwithstanding the potential costs associated with the risk of having
piles that are too short, making the excess length, cutting off a pile to the needed
elevation, and disposing of the remains, are real costs. Is there a way to obtain better
soils data, e.g., through real-time data collection and feedback on how deep piles are
driven into the ground, that may assist in determining cut-off lengths?
4. Are there product alternatives on the market that are more in line with lean
principles? E.g., one overseas company is making high-strength spun concrete piles
of standard-length sections, ending in metal plates so that sections can be welded
together. Is this a viable alternative to the precast concrete piles used on this project?
CONCLUSIONS
This paper presented a study of precast pile delivery and highlighted various kinds of
variability and uncertainty in this process. The description focused on handoffs between
industry participants in the delivery process, lead times, and inventory buildup. The paper
cited several lean construction principles and suggested how they could be applied to this
process with characteristically high demand variability. Additional research is in order to
further articulate the factors that affect this process and that can be altered through
management decisions with regards to production system design.
ACKNOWLEDGMENTS
The authors much appreciate contributions made by industry practitioners, consulted in
the course of this research, who made time to explain their practices and reasoning, and
to entertain alternative ideas. We specially want to thank Cahill in San Francisco, KieCon in Antioch, Treadwell&Rollo in San Francisco, American Piledriving in Pleasanton,
and UC Berkeley for providing the opportunity and facilities to conduct this research.
REFERENCES
Arbulu, R.J. and Tommelein, I.D. (2002). “Alternative Supply-Chain Configurations for
Engineered or Catalogued Made-To-Order Components: Case Study on Pipe
Supports used in Power Plants.” Proc. 10th Annual Conference of the International
Group for Lean Construction (IGLC-10), Gramado, Brazil, August 6-8, pp. 197-209.

9


Ballard, G. (1999). “Improving Work Flow Reliability.” Proc. 7th Annual Conf. Intl.
Group for Lean Construction, IGLC-7, Berkeley, CA, July 26-28, pp. 275-286
Ballard, G. (2001). “Malling Precast Products Ltd.: A Case Study in the Application of
Lean Production Concepts,” Tech. Rept. 2001-1, Lean Constr. Institute, Ketchum, ID.
Ballard, G. and Arbulu, R.J. (2004). “Making Prefabrication Lean.” Proc. 12th Ann. Conf.
Intl. Group for Lean Construction, Elsinor, Denmark, see http://www.iglc2004.dk/
Ballard, G., Harper, N., Zabelle, T. (2003). “Learning to See Work Flow: Application of
Lean Production Concepts to Precast Concrete Fabrication.” Engineering,
Construction, and Architectural Management, Blackwell Pub., 10 (1) 6-14.
Ballard, G., Tommelein, I., Koskela, L., and Howell, G. (2002). “Lean Construction
Tools and Techniques.” Chapter 15 in R. Best and G. de Valence (editors, 2002).
Design and Construction: Building in Value. Butterworth-Heinemann, pp. 227-255.
Brown, R.W. (2001). Practical foundation engineering handbook. Second edition,
McGraw-Hill, page 4.102.
Elfving, J.A. (2003). Exploration of Opportunities to Reduce Lead Times for Engineeredto-Order Products. Ph.D. Diss., Dept. of Civil & Envir. Engrg., University of
California, Berkeley, Fall, 322 pages.
Gil, N., Tommelein, I.D., Kirkendall, R.L., and Ballard, G. (2001) “Leveraging
Specialty-Contractor Knowledge in Design-Build Organizations.” Engineering,
Construction, and Architectural Management (ECAM), 8 (5/6) 355-367.
Hopp, W.J. and Spearman, M.L. (2000) Factory Physics. Second Edition. McGraw-Hill
International Editions, Boston, 698 pp.
Howell, G.A. and Ballard, G. (1999). “Design of Construction Operations.” LCI White
Paper-4, http://www.leanconstruction.org/pdf/WP4-OperationsDesign.pdf, 9 pages.
Howell, G.A., Laufer, A., and Ballard, G. (1993) “Interaction between Sub-cycles: One
Key to Improved Methods.” J. of Constr. Engrg. and Mgmt., ASCE, 119 (4) 714-728.
Koskela, L., Ballard, G., and Tanhuanpaa, V.P., (1997) “Towards Lean Design
Management.” Proc. 5th Annual Conference of the International Group for Lean
Construction, Griffith University, Gold Coast, Australia, July.
Koskela, L., Howell, G., Ballard, G., and Tommelein, I. (2002). “The Foundations of
Lean Construction.” Chapter 14 in R. Best and G. de Valence (editors, 2002). Design
and Construction: Building in Value. Butterworth-Heinemann, pp. 211-226.
Mikulina, T.W. (1998). “The Parallel between Product Development and Production and
Building Design and Construction.” DBIA Conf., Chicago, IL.
Napier (2004). Pile Foundation Design: A Student Guide, Napier University, Edinburgh,
http://sbe.napier.ac.uk/projects/piledesign/guide/chapter2.htm, visited on Sept. 29.
Peurifoy, R.L., Ledbetter, W.B., and Schexnayder, C.J. (1996). “Piles and Pile-Driving
Equipment.” Ch. 15 in Construction Planning, Equipment, and Methods. McGraw
Hill, 5th ed., 633pp.
Tommelein, I.D. (1998). “Pull-driven Scheduling for Pipe-Spool Installation: Simulation
of Lean Construction Technique.” ASCE, J. Constr. Engrg. Mgmt., 124 (4) 279-288.
Treadwell&Rollo (2002). Geotechnical Investigation Block N3A Parcel 1 Mission Bay,
San Francisco, CA. Treadwell&Rollo, San Francisco, CA, 22 May, 34 pp. and appx.
Walsh, K.D., Hershauer, J.C., Tommelein, I.D., and Walsh, T.A. (2004). “Strategic
Positioning of Inventory to Match Demand in a Capital Projects Supply Chain.”
ASCE, J. of Constr. Engrg. and Mgmt., 130 (6) 818-826.

10

Kaplan05analysis

More Related Content

What's hot (19)

Viewers also liked (17)

Similar to Kaplan05analysis (20)

More from Junaida Wally (11)

Recently uploaded (20)

Kaplan05analysis