238
Accessing Geoenvironmental Data: From On-line to Expert Systems
1
INTRODUCTION
A geotechnical investigation is done by means of drilling, sampling, in-situ and
laboratory tests. This investigation provides qualitative and quantitative
information regarding specific locations on the site where soil information,
borings and/or tests are available. This common practice may only cover a small
fraction of the area of the site under investigation, which introduces variable
uncertainty in the soil characteristics. But this uncertainty is unavoidable from an
investigation-cost point of view. Another important source of uncertainty is the
bias in the measuring procedures.
Geotechnical tests involve examining a given soil with a specific device according
to a standardized procedure. Response of a given soil to a particular test depends,
in ideal execution conditions, on the soil type, as well as other factors including
initial state of stress and test type. Unfortunately, the ideal execution conditions
may not be well defined and may considerably vary according to the soil type.
Data obtained from geotechnical tests have to be analysed and assessed for quality
before use in any formal design in order to take into account the effect of the
influencing procedural factors.
2
BASIC ASSUMPTIONS IN CHECKING QUALITY
PROCESS
Geotechnical tests are divided into two main groups: laboratory tests and in situ
tests. Laboratory tests can be divided into two main classes: identification tests
performed on disturbed and "assumed" to be representative samples, and
mechanical tests performed on what are "assumed" to be undisturbed samples. The
basic theoretical assumption made for interpreting the results of a mechanical
laboratory test is that the stress-strain field is homogeneous, or in other words, the
stresses and strains can be measured at a particular point then assumed to be valid
across the whole specimen. Interpretation of mechanical laboratory tests is done
with the assumption that the disturbance during sampling, storing, preparing and
testing samples has been highly reduced.
In situ tests are performed at the site either by pushing a testing device into the
soil, or by pre-drilling a borehole before the introduction of the testing device. In
either case, interpretation of the results is made assuming that a minimum amount
of soil disturbance occurs during the test. In most cases, interpretation of in-situ
tests is done by empirical methods developed and verified through experience.
Other than identification tests which involve simple, direct and systematic
procedures, results of mechanical laboratory and in situ tests are highly subjective
and depend on the quality of the test performed.
3 DATA QUALITY ANALYSIS
Checking the quality of any geotechnical test data involves verifying the validity
of the basic assumptions made for interpreting the test which produces the data. If,
in the triaxial test, a localization of deformations on a specific plane occurs, the
continuity assumption of the stress-strain field becomes invalid.
Accessing Geoenvironmental Data: From On-line to Expert Systems
1
INTRODUCTION
A geotechnical investigation is done by means of drilling, sampling, in-situ and
laboratory tests. This investigation provides qualitative and quantitative
information regarding specific locations on the site where soil information,
borings and/or tests are available. This common practice may only cover a small
fraction of the area of the site under investigation, which introduces variable
uncertainty in the soil characteristics. But this uncertainty is unavoidable from an
investigation-cost point of view. Another important source of uncertainty is the
bias in the measuring procedures.
Geotechnical tests involve examining a given soil with a specific device according
to a standardized procedure. Response of a given soil to a particular test depends,
in ideal execution conditions, on the soil type, as well as other factors including
initial state of stress and test type. Unfortunately, the ideal execution conditions
may not be well defined and may considerably vary according to the soil type.
Data obtained from geotechnical tests have to be analysed and assessed for quality
before use in any formal design in order to take into account the effect of the
influencing procedural factors.
2
BASIC ASSUMPTIONS IN CHECKING QUALITY
PROCESS
Geotechnical tests are divided into two main groups: laboratory tests and in situ
tests. Laboratory tests can be divided into two main classes: identification tests
performed on disturbed and "assumed" to be representative samples, and
mechanical tests performed on what are "assumed" to be undisturbed samples. The
basic theoretical assumption made for interpreting the results of a mechanical
laboratory test is that the stress-strain field is homogeneous, or in other words, the
stresses and strains can be measured at a particular point then assumed to be valid
across the whole specimen. Interpretation of mechanical laboratory tests is done
with the assumption that the disturbance during sampling, storing, preparing and
testing samples has been highly reduced.
In situ tests are performed at the site either by pushing a testing device into the
soil, or by pre-drilling a borehole before the introduction of the testing device. In
either case, interpretation of the results is made assuming that a minimum amount
of soil disturbance occurs during the test. In most cases, interpretation of in-situ
tests is done by empirical methods developed and verified through experience.
Other than identification tests which involve simple, direct and systematic
procedures, results of mechanical laboratory and in situ tests are highly subjective
and depend on the quality of the test performed.
3 DATA QUALITY ANALYSIS
Checking the quality of any geotechnical test data involves verifying the validity
of the basic assumptions made for interpreting the test which produces the data. If,
in the triaxial test, a localization of deformations on a specific plane occurs, the
continuity assumption of the stress-strain field becomes invalid.