As introduction to the mentioned LRFD (U.S.) oder partial safety factor concept (EU) I recommend:

Robert E. Melchers: Structural Reliability - Analysis and Prediction (U.S.)

Gulvanessian, Calgaro, Holicky: Designers' Guide to EN 1990, Eurocode: Basis of structural design (EU)

Hi Jaiwei,

The information which  you are seeking is noted in all codes. But if you wish to know why, then you need to learn a process known as the code calibration.

For a simple answer read these

http://www.steel-stainless.org/Content/Files/papers/35_Baddoo_and_Franci.pdf

http://geo.citg.tudelft.nl/workshop_delft_homepage/workshop_delft_handout.pdf

If you wish to dig deeper read these:

For ISO codes

http://www.ogp.org.uk/pubs/422.pdf

http://www.hse.gov.uk/research/rrpdf/rr087.pdf

http://www.hse.gov.uk/research/rrpdf/rr088.pdf

For bridges

http://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_rpt_454.pdf

For Eurocodes

http://www.km.fgg.uni-lj.si/coste24/data/jccs/pdf/gulvanessian_holicky.pdf

http://www.efca.info/downloads/Commentary_to_Eurocode.pdf

These are just a few samples.  Search the Researchgate.net or the internet using keywords code calibration, partial safety factor and the code’s name.

If you find it too much to read, then narrow your question by specifying a code, which material (wood, steel, concrete,..)  and how are going to use the data.

Good luck for your endeavour

Sirous Yasseri

 Thanks for your help, Mr. Achenbach, though I couldn't find these books in our university library. 

Thank you so much Dr. Yasseri. 

There are two types of factors used in the design of reinforced concrete structures. The first is the load factors which is usually > 1.0 and based on probabilistic considerations. The second type of factors is the strength reduction factor which is usually < 1.0 and depend on:

1- Type of failure,

2- Importance of member,

3- Accuracy of the used design equations,

4- Level of supervision on the site and how the design drawings can be performed like,      placement of reinforcement, dimensions of the members, curing of concrete etc.

The nominal strength values of strength like, axial, moment, shear, torsion should be reduced by this strength factor to get a design strength.

For example, the strength reduction factor equal to:

0.65 to 0.9 for flexure or bending moment and axial load (tied members),

0.75 to 0.9 for flexure or bending moment and axial load (spirally reinforced members)

0.75 for shear and torsion.

For mix design there are other probabilistic values used in the mix preparation. 

Hi Jiawei,

If you don't have access to Melcher's book, then these are good references

http://www.kstr.lth.se/fileadmin/kstr/pdf_files/STforsk_kurs-10/John_D-S_lecture_notes.pdf

http://e-collection.library.ethz.ch/eserv/eth:25307/eth-25307-01.pdf

http://civile.utcb.ro/ccba/srracourse.pdf

http://www.km.fgg.uni-lj.si/coste24/data/coimbradocuments/coimbra-faber.pdf

You need to read them in the same order as above.

I hope it helps

Sirous Yasseri

Thanks, Prof. Altaan. These safety factors have incorporated some considerations of failure probability, right? But are the bias factors related to some certain level of exceedance probability? If so, there must be some statistics from which the probability distributions of the variables (e.g., the strength of the materials) are obtained. Where can we find these statistics?

Thanks a lot Dr. Yasseri. I can see that these are very good resources. I will try to find the answers to the above questions from them.

Hi Jiawei

It seems you are looking for data and see how the statistical distribution of of various parameters, governing the design, are determined. Links in my first communication give you some information, not all of it.  the following link  a paper Fred Moses  explain the concept of using reliability to calibrate API code. If you can find his earlier papers cited in that paper, you will find more detail, but all of them are older than you.

http://www.shipstructure.org/pdf/87symp02.pdf

You will find some answer to your question in this document.

http://onlinepubs.trb.org/onlinepubs/circulars/ec079.pdf

Partial safety factors (both load and resistance) have been derived based on  a mixture of fresh data and their statistical distribution, and previous codes in existence. Partial safety factor were originally set for the modern code so that results to be similar to the old code. Such parity has gradually been eroded in the later version of codes. 

Regards

Sirous Yasseri

Thanks Dr Saad for information.

valuable information from dr. saad

Jiawei Chen ...Please check references below.. hope it will be helpful for you…

Regards…

https://www.ce.jhu.edu/bschafer/reliability/Buonopane_SSRC.pdf

http://www.awarie.zut.edu.pl/files/ab2017/referaty/01/01-04%20-%20Nowak%20A%20S%20,%20Iatsko%20O%20-%20Probability%20of%20failure%20%E2%80%93%20load%20and%20resistance%20factors.pdf