Predicted & Measured Behavior of Five Spread Footings
Predicted & Measured Behavior of Five Spread Footings
(OP)
Many of you may have already read ASCE Geotehnical Special Publication No. 41, "Predicted and Measured Behavior of Five Spread Footings on Sand", ASCE, 1994. I am just finishing it and find it very interesting. The Federal Highway Admin. and the University of Texas sponsored the study.
A series of soil tests, Borehole Shear Tests, Cross-Hole Wave Tests, PiezoCone Penetration Tests, Dilatometer Tests, Step Blade Tests, Standard Penetration Tests, and five others where performed on the same site. The test data was sent out to respondants who volunteered to predict load and settlements for Q25, Q150, s @ 30 min. and s @ 20 years.
Five footings were constructed and tested and then compared to the prior predictions of 31 engineers (some were teams of 2-3 engineers). The predictors could use whatever methods they wished, but all had to report based on the information provided and in a standard format.
I found the reading quite interesting and the results somewhat surprising. A prediction was considered successful if it was within +/-20% of the actual test values. No one had ALL their answers fall with the +/-20%. But, the description of the participants methods and their reasoning is pretty educational. If you haven't read it, I think it is well worth the effort and the $32 price (less if you are a member of ASCE).
A series of soil tests, Borehole Shear Tests, Cross-Hole Wave Tests, PiezoCone Penetration Tests, Dilatometer Tests, Step Blade Tests, Standard Penetration Tests, and five others where performed on the same site. The test data was sent out to respondants who volunteered to predict load and settlements for Q25, Q150, s @ 30 min. and s @ 20 years.
Five footings were constructed and tested and then compared to the prior predictions of 31 engineers (some were teams of 2-3 engineers). The predictors could use whatever methods they wished, but all had to report based on the information provided and in a standard format.
I found the reading quite interesting and the results somewhat surprising. A prediction was considered successful if it was within +/-20% of the actual test values. No one had ALL their answers fall with the +/-20%. But, the description of the participants methods and their reasoning is pretty educational. If you haven't read it, I think it is well worth the effort and the $32 price (less if you are a member of ASCE).





RE: Predicted & Measured Behavior of Five Spread Footings
1. My impression is that the more complicated the methods used, the less accurate! This is only my impression at this point, but maybe that old saying has a grain of truth, we can be "educated beyond our intelligence"! If +/- 20% is the "target", maybe FEM and lots of differential equations are not needed?
2. It was interesting to go through each respondant's list of references. There were many that were the same, Bowles, Schmertmann, Terzaghi & Peck and Vesic seemed to be the most popular. But, there is a wealth of other references too.
RE: Predicted & Measured Behavior of Five Spread Footings
Short term my experience is that values furnished by the books for the situation use to be very conservative. Long term may be not as much, but may also be.