FOUNDATION ANALYSIS COMPUTER PROGRAMS BASIC INFORMATION
What Are Foundation Analysis Computer Programs?
A large
number of foundation engineering problems can be efficiently analyzed and/or
designed using a digital computer. Particular advantages of using a computer
accrue from these features:
1. One is
able to try a range of problem variables to obtain a feel for the effect of
specifying, or using, a particular set of soil parameters.
2. One can
avoid having to use tabulated data or plotted curves, which usually require
interpolation and excessive simplification of the foundation model.
3. One can
minimize computational errors from these sources:
a. Erroneous
key entry when using a calculator. The bad entry is (or should be) output to paper
using a computer so the input can be checked.
b. Omission
of computational steps. A working computer program usually includes all the design
steps. A set of hand computations may not include every step for any number of
reasons (forget, not be aware of, carelessness, etc.).
c.
Calculator chip malfunction not readily detected except by using two
calculators. Computer chips are often internally checked on power-up, or output
is so bad that chip errors are visually detected.
4. With
output to a printer one has a paper record of the problem for office files
without the necessity of transcribing data from intermediate steps. This avoids
copy errors such as 83 for 38 and the like.
The major
disadvantage of using a computer program is that it is difficult to write a
first generation, error-free program of real use in a design office. Program
usability tends to increase with each revision (or history) level.
With the
current wide availability of computer programs—many, such as those on the included
diskette, having a "history"—the advantages gained from program use
far exceed any perceived disadvantages.
The author
suggests that both geotechnical and foundation engineers should use computer
programs whenever possible—and certainly be aware of what computer program(s)
each is likely to use for the given project.
This
statement is made with full awareness of the possibility of program errors (or
"bugs"). Fortunately, most geotechnical software is task-specific so that
the possibility of program errors or their not being detected is not so likely
as for some of the large finite-element or structural analysis programs that
purport to solve a wide range of tasks.
In any case,
the author cannot recall a single reported foundation design failure that can
be attributed to a bad4 computer program. It should be evident that computer
programs vary widely in perceived quality, perceived quality being defined here
as problem limitations and "ease of use." Both users and programmers should
be aware that it is difficult to predefine the full range of problem parameters
likely to be encountered in practice, so nearly any geotechnical program of
significant value is likely to have some built-in limitations.
Ease of use
is highly subjective and depends more on user familiarity with a program than how
easy it really is to use—many users like pulldown menus and graphics whereas
others are quite content without these features. As a final comment on computer
programs, be aware that although business applications and games usually have a
market in the hundreds of thousands, geotechnical programs have a potential market
of only a few thousand.
This small
market means geotechnical software is likely to be more expensive than other
software and, to minimize development costs, it is not likely to have many so-called
user-friendly features.
One should
routinely check the output from any computer program used for design or analysis.
The user is responsible for his or her design since it is impossible to write a
computer program with any usefulness that cannot be misused in some manner. Primarily
for this reason most computer programs are sold or licensed with a disclaimer
making the user responsible.
Fortunately,
most computer programs can be written to be somewhat self-checking, either by
writing back the input data or by providing output that can be readily
identified as correct (or incorrect) if the user understands or knows how to
use the program. It should go without saying that, if you do not know much
about the specific problem being designed or analyzed, you should first do some
preliminary study before using a computer program on it.
