In May of 2000, I participated in an AEC conference
in San Francisco, during which I had the opportunity to visit
with a variety of industry practitioners and dot.com executives.
One could unmistakably sense the fear throughout the speakers
and attendees alike. On the one hand, practitioners worried
over their lack of understanding of various Internet-enabled
technologies, which would supposedly redefine their businesses,
if not eliminate them entirely. On the other, dot.com executives
knew that, without achieving sufficient market penetration and
scale within short investment horizons – as defined by
venture capitalist expectations – they would be simply
an historical footnote.
However, everything changed on April 14, 2001
as tech stocks disintegrated and while many of the AEC practitioners
are still alive, most of the dot.coms serving the AEC industry
have failed, burning through, by some estimates, over $1.3 billion
in capital. In most cases, their value propositions were insufficient
to drive economic efficiency throughout a highly fragmented,
low margin industry which still remains highly wasteful in its
delivery process. They relied upon business models which exploited
new internet technologies to enable faster communication and
better organization of information – a much needed contribution
to the industry, but not of significant value. The speed of
communication, with better organized information, is not the
critical issue. Just because the pipeline gets bigger, permitting
more uncoordinated information to flow through it faster, does
not mean that the value gained within the process has increased
significantly, nor are the individual participants necessarily
any better off. In fact, by improving the efficiency of communication
between the disciplines, designers and engineers are potentially
motivated to provide less complete and coordinated information
to contractors, knowing that issues can be resolved later using
sophisticated communication technology with subcontractors in
the field – precisely the point where resolution is the
most expensive. Why would designers absorb the cost of fully
developing their drawings in advance, only to revise their documents
during construction when more precise information becomes available?
There has been much written about how the Internet
might redefine organizations in the future, including our definition
of “work.” However, it is increasingly apparent
that the Internet is only the most prominent facet of an “Information
Revolution” in which the costs of computing, data storage,
and communication are rapidly approaching zero, due largely
to both technological innovations and a tremendous investment
in fiber capacity during the boom of the 1990’s. This
“revolution” is radically changing where, when,
and how we work.
One might compare the information revolution to
the impact of technologies in a prior age when rail, steam,
and, ultimately, electricity redefined manufacturing and reshaped
the “organization.” As transportation costs declined,
manufacturers increasingly located their plants near the best
available labor sources. Electricity afforded manufacturers
the opportunity to optimize the location of equipment rather
than having to centralize around a waterwheel. The same kind
of redefinition of where, when, and how we work is occurring
today, as talent chooses where it wants to live (regardless
of “the place of employment”), when it wants to
work (24/7), and how it chooses to do so, (utilizing sophisticated
computing, data storage, and communications networks). This
phenomenon is driving the growth of “ex-urbia” in
America where talent can live, work, and play without wasting
time commuting. The rise of the knowledge-based economy has
caused signifiicant transformations, but we still struggle with
the question: “What will the AEC organization of the future
look like?”
The Barriers to Change
Before suggesting what the future might hold
for the industry, we must first try to understand its challenges
today. Engineering News Record, the Construction Industry
Institute, the Lean Construction Institute, and a variety of
other industry publications and associations have documented
much data substantiating the magnitude of waste inherent in
the traditional AEC delivery process. We’re not speaking
of one to five percent issues here, but rather a twenty to forty
percent waste in the time and cost required to deliver projects
using conventional methods. For a more detailed discussion of
the causes of such waste and the barriers to eliminating them,
please refer to "The AEC Dilemma: Exploring the Barriers
to Change”, Design Intelligence, Focus on the Future.
March, 2001; www.di.net.
A fragmented industry, low margins precluding meaningful investment,
a lack of integration in a zero-sum contractual model, and optimizing
returns within the silos of each discipline despite the impact
on project costs, are a few of the many barriers to removing
waste within the industry.
Perhaps the greatest barrier to change, however,
results from our motivations as participants. We have all been
rationally trained to believe that if something can be done
better, then the improvement should and will be implemented.
Nothing could be farther from the truth. As Steve Hindman so
effectively articulated in his article, “Hitting A Nerve”
(Context Magazine, July 2002), it is the painful nerve
within a system that gets tended to first, regardless of the
magnitude of the benefit derived from any other changes. In
other words, just because something is better, faster, and/or
cheaper doesn’t mean it will be implemented, unless an
individual or commonly motivated group of individuals (the enterprise)
realizes a direct benefit.
Consequently, improvements in a fragmented industry
are slow to come and particularly difficult to implement. Witness
the retail industry in the 1970’s. Independent, small
retailers were reluctant to invest in credit card reading technology
which would have substantially reduced transaction costs, across
the entire system, by eliminating the need for checks; banks,
not retailers, would have realized most of the benefits. It
wasn’t until the implementation of bar code technology,
which reduces inventory-tracking costs at the retail level,
did retailers incorporate the card reading technology that dramatically
reduced transaction costs throughout the entire payment system.
The AEC industry offers many such examples of
potential industry advances that are never realized because
the “investor” receives no benefit. One such example
is the time and cost associated with completing the project
design before construction commences, thereby avoiding substantial
costs associated with rework in the field. Architects, general
contractors (GC’s), subcontractors, and fabricators collectively
devote considerable man-hours passing shop drawings and other
forms of poorly coordinated submittals back and forth, adding
significant time and costs to both the design and construction
of a project. This dysfunctional and costly process is partially
a result of 1) the customization of components to the nth degree,
2) insufficient tools to intelligently and accurately model
buildings in three dimensions, 3) inadequate design and engineering
fees to cover the cost of completing the design before construction
commences, and 4) the lack of clarity within the supply-chain,
particularly between designers and fabricators/subcontractors
– the latter having the detailed information about the
building components, while the former specifies much of the
design.
This dilemma is rooted in the continued adoption
of historical practices which limit opportunities for significant
improvements. When buildings were relatively simple, a design-bid-build
system seemed to work fairly well. The move to this approach
was due, in part, to the push for specialization arising from
the Industrial Revolution. Both bidding and increased design
complexity drove the industry towards specialization within
the design disciplines and the trades. Fifty years ago, GCs
self-performed much of their work. Today, most such work is
subcontracted to specialists, with the lowest bid, even in a
collaborative environment in which the GC and architect work
together from the outset. Most often, it is the subcontractor
or fabricator who knows the most about the design details, but
they are rarely engaged early enough in design to have much
impact because the bidding process requires partially completed
documents by which to compare bidders. Amazingly, many owners
are unaware of the waste of their investment dollars simply
because it is so hard to measure. In addition, much of the industry
has a vested interest in current practices. Most owners who
do understand the problem are reluctant to fight the battle
with their purchasing or audit departments, the latter of which
are rarely encouraged to experiment with alternative contractual
models which might motivate participants to improve any aspect
of the delivery process.
Even greater waste in cost and time occurs, after
the bidding process, during the actual construction of the building.
For example, the cost of fire suppression systems could be reduced
by at least thirty percent if the subcontractor knew the final
location of the lighting and HVAC distribution systems before
performing the work. But, all of the benefits derived through
a bid process would pass on to the project owner. Designers
and engineers could not possibly afford to sufficiently coordinate
the lighting and HVAC designs in advance of bidding the fire
sprinkler system, so the design is completed by the sprinkler
contractor during construction with many on-site changes –
a highly wasteful practice. Similar waste occurs in many other
trades for similar reasons.
The bottom-line is that this industry is slow
to adopt innovative practices because participants are not rewarded
for doing so in a high-risk environment. It is not a function
of a lack of talent or some errant behavior which make builders
and owners change-averse, as some might imply. Owner’s
representatives fail because projects run over, but they can
hardly be called heroes for bringing projects in on time and
within budget, since that is what is expected of them. As we
learned from our retail example in the 1970’s, just because
something is better, faster, or cheaper from a total-system
viewpoint does not mean that the industry will adopt it.
Collaboration vs. Integration
All work processes usually involve some combination
of four types of “work” which can be though about
as a continuum. At one end of the continuum is the entirely
independent effort of the individual to produce a product or
service as pioneers built homes using tools they fashioned themselves
from natural resources. The next type consists of dependent
work wherein one applies some level of effort using someone
else’s product or service much like one would use milled
lumber to manufacture furniture. Further along the continuum,
participants must interactively share information across disciplines
in order to achieve their mutual objective. At this level, no
one person or entity contains sufficient knowledge to independently
accomplish a complex task. This is directly analogous to collaboration
between team members in the AEC environment, where information
is shared across disciplines throughout the delivery process.
Finally, at the other end of the continuum, there is the integration
of knowledge acquired through specialists in a range of disciplines
enabling the team to optimize the work to be performed.
Let’s focus on those activities which are
dominated by either collaboration or integration – two
distinctly different concepts which can occur simultaneously
but are often misused interchangeably. Collaboration
is a data-centric activity wherein each discipline contributes
data information to other disciplines for processing to achieve
common objectives. In the AEC world, we refer to this as the
team approach wherein a negotiated contractual arrangement is
used to engage the GC, architect, and engineer early in the
design process, with the hope of designing the best building
for the budget. Information is passed back and forth between
disciplines with little concurrent processing based on shared
knowledge. A great example of collaboration is an engineer designing
the building structure based on a 30-day old set of architectural
drawings while the architect continues making design changes,
thus causing coordination problems and costly rework during
fabrication and construction.
By contrast, integration is a knowledge-centric
activity wherein each discipline contributes knowledge in the
form of rules, algorithms, and proprietary practices –
an approach not followed since the master builder dominated
the industry during the 19th century, when building design was
substantially simpler and one person could hold most of the
necessary knowledge in their head. Unlike collaboration, integration
relies on participants sharing their knowledge, perhaps by encoding
it in object-based technologies to model the building (commonly
referred to as the Building Information Model, or BIM) –
a daunting thought to the many practitioners whose livelihoods
depend upon their knowledge of a discipline.
One of the fundamental differences between data-centric
and knowledge-centric environments is that the former requires
identification of a specific project before work can begin,
applying the knowledge within disciplines and sharing data between
them, while an integrated environment entails the sharing and
encoding of rules, algorithms, and other proprietary knowledge
and practices before any project is considered. In an integrated
environment, once a project is identified, only project specific
data must be processed, thereby significantly reducing processing
time and costs while substantially improving the coordination
and completeness of design information and radically reducing
waste in the field. With new, and better, design tools, such
an integrated approach offers the opportunity to evaluate many
more options in a virtual environment, at far lower costs, before
irrevocably committing within a physical environment
Consider one of many simple examples. Structural
engineers in a collaborative environment design each concrete
or steel beam in a building only after the architect
has provided a substantial amount of project specific design
information about the expected use, building outline, layouts,
etc. Any thorough value-engineering effort to reduce costs through
significant redesign is prohibitively expensive, even if sufficient
time were available. However, using an integrated approach,
the engineering formulas, code requirements, and other rules
which influence the design of structural beams are embedded
within intelligent objects in advance of any project being considered,
much like the technology used in the automotive and aviation
manufacturing industries. After a project is identified, only
project-specific information is inserted to create a unique
model. In addition, multiple evaluations of the model are easily
performed, in a short period of time and at little cost, by
adjusting the project-specific criteria to evaluate the effects
on structural design, building costs, etc. Again, the primary
difference between approaches is that collaboration entails
the constant reuse of rules, etc., by professionals for each
unique project while integration encodes the professional’s
knowledge within a rule-driven tool which others may modify
to evaluate and optimize their own design input much more efficiently.
Driving Change within the AEC Industry
Let’s look at some remarkable results derived
from integrating knowledge across various disciplines and/or
professions. Are there some common characteristics in the AEC
industry from which we can learn?
Up until the recent Afghan war, US bombers were assigned objectives
by spotters informing Pentagon attack planners of potential
targets. After evaluation, these attack planners transmitted
orders to planes which then required several hours to reach
their targets. This worked great for stationary targets, but
mobile targets disappeared long before the planes could spot
them. Remember, during the initial stages of the Afghan war,
political pressure on the Department of Defense grew rapidly
as the bombing campaigns appeared to be ineffective against
a highly mobile enemy. The targeting process quickly evolved
to spotters, using wireless technology, communicating directly
with pilots who maintained their flights over periods of time
in target-rich zones (Business 2.0, Oct. 2002). Moving
vehicles now became viable targets and results improved dramatically.
Off the record, some Air Force officials suggested that the
new process would not be used again because of uncontrolled
access to data – or were they really worried about their
future roles in the process?
A floundering gold company in the Red Lake district
of Ontario, as a last ditch effort to save their gold mining
operations, asked for recommendations from mining experts world-wide.
By sharing their (proprietary) metallurgical data on the web,
over the objections of their geologists, the company implemented
the best recommendation, increasing production nearly ten times
while reducing costs by 85 percent. (Fast Company, June
2002) Interestingly, the winning suggestion was submitted by
a 3-D modeling consortium which had the technology to evaluate
the data more effectively, as object-based design tools may
do for the AEC industry.
Along with the retail industry’s check processing
example referred to earlier, what do these process improvements
(and so many others) hold in common? First, there usually exists
a particularly inefficient or wasteful process upon which all
participants rely to accomplish a complex task. Secondly, a
new tool (often referred to as a technology, until it becomes
widely accepted) is introduced which enables the process to
be improved. Finally, there exists a driver for change resulting
from some participant’s intense pain which motivates that
participant to modify their behavior and dispense with accepted
rules. (For a much richer understanding of some of these ideas,
refer to The Only Sustainable Edge by John Hagel III
and John Seely Brown; Harvard Business School Press.). Are these
three characteristics inherent in the AEC industry today?
Clearly, the industry is plagued by inefficient
and wasteful processes, and many participants are feeling the
pain. Every discipline is being “commoditized” resulting
in anemic margins, while customers are often frustrated with
unexpected outcomes. The necessary technologies are increasingly
available to radically improve current processes by integrating
knowledge through encoded rules and algorithms – Revit,
Archicad, and MicroStation are emerging examples of such tools
and interest in these technologies is unquestionably growing.
Most importantly, practitioners increasingly realize that some
better form of knowledge integration between the disciplines
will be required to use these tools. So what are we missing?
The driver! As potential alternative delivery
methods are better understood, owners will insist that processes
be improved to achieve increasingly possible results. Already
some large users, like the GSA and British Airport Authority,
are insisting on some form of integration including a common
Building Information Model for all to develop and share. As
the data increasingly warrants better designs at lower costs
with faster schedules, we can expect other customers to demand
similar results derived from shared knowledge. The only remaining
questions are: how long will this revolution take; and to what
extent will AEC participants help mold it for the benefit of
all instead of avoiding the issue.
The answer to the first question requires a crystal
ball. But the second is up to us. As practitioners, we need
to lead this revolution instead of letting construction managers,
brokers, and the like define our future for us. We must start
by admitting to ourselves that we do waste significant time
and money employing conventional methods. Second, we must put
aside our fear of the unknown and experiment with new risks,
emerging technologies, alternative business models, and unfamiliar
new roles. As architects, engineers, contractors, etc., we must
jointly share both the authority and responsibility for design,
means and methods, costs, and delivery schedules by sharing
contract risks on a project-by-project basis and/or integrating
our respective disciplines within the same firm.
Most of us are reluctant to do so. We can’t,
then, complain about becoming a commodity or losing control
of the process. This is all about regaining control (thereby
reducing risk), improving margins, and contributing to a better
built environment which we owe to our clients, our communities,
and ourselves. Yes, many of our clients are not asking for it,
but then I don’t know anyone who asked for the fax machine,
computer, or the cell phone. That crystal ball may reflect a
future much closer than we think – so what’s your
choice?
* * * * * THE END * * * * *
