Justi cation for Improvement Vertical Cut in Cohesive Soil

Reading UCC - 12 In VS .NETUsing Barcode Control SDK for Visual Studio .NET Control to generate, create, read, scan barcode image in .NET framework applications.

Example 13.1 described the reliability analysis of a vertical cut in cohesive soil expressed in terms of the margin of safety M M = c H /4

UPC A Generation In VS .NETUsing Barcode generation for .NET framework Control to generate, create UCC - 12 image in .NET applications.

Reliability and Statistics in Geotechnical Engineering Gregory B. Baecher and John T. Christian 2003 John Wiley & Sons, Ltd ISBN: 0-471-49833-5

Scan UCC - 12 In .NET FrameworkUsing Barcode recognizer for VS .NET Control to read, scan read, scan image in Visual Studio .NET applications.

(16.1)

Create Bar Code In .NET FrameworkUsing Barcode generation for .NET Control to generate, create bar code image in VS .NET applications.

THE HASOFER LIND APPROACH (FORM)

Decoding Bar Code In VS .NETUsing Barcode decoder for .NET Control to read, scan read, scan image in .NET framework applications.

where c is the cohesion, is the unit weight, and H is the height of the cut. For a 10 m deep cut and uncorrelated cohesion and unit weight, the following values were used: H = 10 m c = 100 kPa = 20 kN/m3 Then c = 30 kPa (16.2)

Encode GTIN - 12 In Visual C#Using Barcode generator for Visual Studio .NET Control to generate, create UPCA image in Visual Studio .NET applications.

= 2 kN/m3

Generate UPC Symbol In .NETUsing Barcode generator for ASP.NET Control to generate, create GTIN - 12 image in ASP.NET applications.

M = 100 (20)(10)/4 = 50 kPa

Paint UCC - 12 In Visual Basic .NETUsing Barcode printer for Visual Studio .NET Control to generate, create UPC-A Supplement 5 image in .NET framework applications.

2 M = (30)2 + (2)2 (10/4)2 = 925(kPa)2 = 50/ 925 = 1.64

Paint Code 128 In .NETUsing Barcode drawer for .NET Control to generate, create Code 128 image in .NET applications.

(16.3)

Make USS-128 In VS .NETUsing Barcode creator for VS .NET Control to generate, create EAN / UCC - 13 image in .NET framework applications.

If the cohesion and unit weight are both Normally distributed, any linear combination of them, such as M, must also be normally distributed, and pf = P [M 0] = ( ) = 5.01 10 2 (16.4)

EAN / UCC - 13 Generator In VS .NETUsing Barcode generation for Visual Studio .NET Control to generate, create GTIN - 13 image in .NET applications.

Note that, because the variables are normally distributed and M is a linear combination of them, these results are exact. Now, let us solve the problem using FOSM. The derivatives are M = 1 and c

Print 2/5 Interleaved In .NETUsing Barcode creation for Visual Studio .NET Control to generate, create 2/5 Interleaved image in Visual Studio .NET applications.

2 2 2 = (1)2 (30)2 +

Code 128 Code Set A Maker In VB.NETUsing Barcode printer for Visual Studio .NET Control to generate, create Code-128 image in VS .NET applications.

M H = 4

Drawing DataMatrix In Visual C#Using Barcode generation for .NET Control to generate, create DataMatrix image in .NET framework applications.

(16.5)

UPC-A Supplement 5 Printer In JavaUsing Barcode creation for Java Control to generate, create UPC A image in Java applications.

The value of M is unchanged at 50 kPa. The FOSM estimate for M is

Making ANSI/AIM Code 128 In JavaUsing Barcode drawer for Java Control to generate, create Code-128 image in Java applications.

2 M

Bar Code Reader In .NETUsing Barcode scanner for .NET framework Control to read, scan read, scan image in .NET framework applications.

M c

GS1 128 Generator In Visual Basic .NETUsing Barcode creation for VS .NET Control to generate, create UCC - 12 image in .NET applications.

c2 +

Paint EAN 13 In JavaUsing Barcode maker for Java Control to generate, create GS1 - 13 image in Java applications.

M

ECC200 Drawer In VS .NETUsing Barcode generation for ASP.NET Control to generate, create ECC200 image in ASP.NET applications.

10 4

(2)2 = 925 kPa

(16.6)

which is identical to the exact result. The rest of the calculation gives once again = 1.64 and pf = 5.01 10 2 . This is what one would expect from a linear combination of Normal variables. However, the problem could also have been stated in terms of the factor of safety F = 4c H and F 4c 1 = H 2 (16.7) (16.8)

F 4 = c H F = 0.020 c F

2 F

If the evaluations are done at the mean values of the variables, then F = 0.100 (16.9) (16.10) F

2 2 = (0.020)2 (30)2 + (0.100)2 (2)2 = 0.4

4 c = 2.00 H F c

c2 +

(16.11) (16.12) (16.13)

= (2.0 1.0)/ 0.4 = 1.58 pf = 3.81 10

JUSTIFICATION FOR IMPROVEMENT VERTICAL CUT IN COHESIVE SOIL

The calculation of the probability of failure assumes F is Normally distributed, which is certainly not the case. However, it is also not clear what the correct distribution of F is. This example shows that, even for a very simple problem, the FOSM procedure gives different results for calculations based on M and on F even though the failure conditions (M = 0 and F = 1) are mathematically identical. Discrepancies and inaccuracies can be expected to become larger as problems become more complicated. Figure 16.1 illustrates the differences between the two computational approaches. Values of c are plotted along the horizontal axis, and values of H /4 are plotted along the vertical axis. The large black dot represents the mean values of both variables. The heavy black line represents the failure condition (M = 0 or F = 1). Points above it and to the left are unsafe, and those below it and to the right are safe. The dashed lines represent constant values of M. The solid lines represent constant values of F . The lines for M = 0 and F = 1 coincide. Reliability analysis is an attempt to quantify how close the system is to failure, that is how close the black dot is to the heavy failure line. The FOSM analysis based on M does this by treating M as a surface whose contours are the dashed lines. Then the most rapid line of descent to M = 0 is the negative gradient of the surface (the positive gradient is uphill). The negative gradient is identi ed on the gure, but, because the contours are parallel, the same gradient would be found anywhere on the M surface. When the same procedure is applied to the F surface, whose contours are the solid lines, the gradient does not represent the shortest distance to the failure condition. Furthermore, different gradients are found at different points. In particular, if we had been able to evaluate the gradient at a point on the line representing the failure condition (F = 1), it would have had the same orientation as the gradient from the M surface and the results from the two approaches would have been the same. The failure to achieve this result is a de ciency of the FOSM approach. Furthermore, even if we did nd the shortest distance from the black dot to the failure surface and if we knew the probability densities of the variables c and , we would not

Figure 16.1 Factor of safety reliability compared to margin of safety reliability. Both methods start from the same point (shown as a large solid dot), but they go in different directions and meet the failure line at different points.

THE HASOFER LIND APPROACH (FORM)

know the distribution of F . The computation of probability of failure requires that we assume that F has one of the common distributions, such as the Normal or LogNormal distributions. The Hasofer Lind approach deals with both these problems.