UNRAVELING
FOUNDATION
UNCERTAINTY Statistical Analysis Offers a Rational Approach for Controlling Foundation
Risk in Design-Build Projects
By Gregory B. Baecher, Ph.D., Ross T. McGillivray, P.E. and Thomas M. Waits, P.E., M.ASCE
For the past 20 years, design-build
(DB) procurement has become
the default for many transportation
agencies for major projects in the
U.S. Although DB offers economic advantages
over design-bid-build, the advantage
can evaporate when the price of major
bridge foundations in the fixed-price
bidding process is based on inadequate
geotechnical information.
DB teams sometimes augment the
geotechnical information presented in
Requests for Proposal (RFPs) by conducting
additional exploration and testing to
reduce uncertainty. However, they usually
have limited time before the bid date, making
such a strategy impractical. What can
DB teams do other than add a contingency
and increase the price of their bid to cover
the risk of uncertain subsurface conditions?
The authors have developed a statistical
decision-making model for predicting foundation
costs and quantifying uncertainty
on future projects. The model is based
on statistical analysis that compares site
characterization information (borings and
tests), geological interpretations, design
assumptions and predictions and pile driving
records from nearby projects, when
available. This method can be expected to
bring value to projects, especially where
subsurface uncertainty is important.
Old (1960) and new (1991) Howard Frankland bridges over Tampa Bay between Tampa to the east
(left) and St. Petersburg to the west (right).jpg
DB foundation challenges for a
new Tampa Bay bridge crossing
There are many locations throughout
the U.S. where high geological variability
impedes DB procurement, such as in
Tampa Bay, Fla. In the past, the subsurface
variability of the Bay has resulted in
pile foundations for bridge support having
enormous variations in depth from
one bridge pier to another, and even
Reprinted with permission from GEOSTRATA (January/February 2020), copyright 2020, American Society of Civil Engineers.
102 | ISSUE 3 2020 www.piledrivers.org
/(right).jp
/www.piledrivers.org
