Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 83
Edited by: B.H.V. Topping, G. Montero and R. Montenegro
Paper 43

Probabilistic Serviceability Assessment of Structures

J. Valihrach1 and P. Marek2

1Department of Structural Mechanics, VSB - Technical University of Ostrava, Czech Republic
2Institute of Theoretical and Applied Mechanics, The Academy of Sciences of the Czech Republic, Prague, Czech Republic

Full Bibliographic Reference for this paper
J. Valihrach, P. Marek, "Probabilistic Serviceability Assessment of Structures", in B.H.V. Topping, G. Montero, R. Montenegro, (Editors), "Proceedings of the Eighth International Conference on Computational Structures Technology", Civil-Comp Press, Stirlingshire, UK, Paper 43, 2006. doi:10.4203/ccp.83.43
Keywords: serviceability, reliability assessment, simulation, Monte Carlo technique, reference values, probability, uncertainty, dynamic analysis.

From the commonly known definition, the serviceability limit states in building structures are conditions in which the functions of the building are disrupted during normal use by excessive deformation, motion, or deterioration.

Nowadays, the codes used in common engineer's practice are based mostly on the "prescriptive" semi-probabilistic partial factor method that was developed and introduced in the pre-computer era. Because of this fact, the method, although based on probabilistic concepts, was subject to inappropriate simplifications. In the case of serviceability assessment, these simplifications include load and load-effect combination rules. Also the expression of serviceability criteria is very simplified by putting more criteria into a simple requirement, thus obscuring the assessment principles from the designer. As it is shown in the examples, this can result in neglecting some of the important serviceability requirements and can often lead to costly reconstructions or modifications of the structure.

In 1986, the ASCE Ad Hoc Committee published a report [1] which focused on issues related to the improvement of serviceability assessment, its interpretation in codes and in designer's work. From some conclusions stated there, it can be deduced that the deterministic approach, based on describing input variables with simple characteristic values and possibly a load factor, is unsatisfactory and that a different approach to the serviceability assessment should be devised.

The introduction and spread of personal computers allowed qualitative improvement of the reliability assessment procedure and the transition to probabilistic concepts. One of the possible approaches to fully probabilistic reliability assessment is the simulation-based reliability assessment method (SBRA). In this method, the attention is given to the engineering reliability, i.e. to the probabilistic assessment of safety, serviceability, and durability. Today, several textbooks [2,3] and more than four hundred published papers, document the potential of the method.

In the framework of the SBRA method, the assessment procedure is based on analysis of reliability function with application of simulation technique. This leads directly to the estimation of the probability of failure which can be compared with the target (design) probability ., that can be determined considering requests submitted by an investor, a code (such as CSN 73 1401 [4]), economic aspects.

When applying the SBRA approach to serviceability assessment, the designer should pay attention to formulation of the serviceability criterion, application of transformation model, definition of the reference value or function, and determination of the target probability. In the examples, particular care is taken to clarify this fully probabilistic serviceability assessment in the framework of SBRA.

The first example addresses a floor vibration problem that occurred due to insufficient natural frequency. The fully probabilistic serviceability assessment is applied to the structure both before and after the modification (based on appending the steel truss girders to the bottom of beams). The results are presented, together with the sensitivity analysis. In the second example a crane track girder unserviceability problem due to the designer's neglect of the required dynamic analysis is concerned. The third example demonstrates one of the first uses of probabilistic serviceability assessment applied to a television tower extension column. The last example is the London Millennium Footbridge, where the serviceability requirements were not initially met.

In the conclusions some recommendations regarding probabilistic assessment of structures and its introduction into design practice are given. Also the direction of future development of the SBRA method is indicated.

Ad Hoc Committee on Serviceability Research, "Structural Serviceability: A Critical Appraisal and Research Needs", Journal of Structural Engineering, ASCE, 112(12), 2646-2664, 1986. doi:10.1061/(ASCE)0733-9445(1986)112:12(2646)
P. Marek, M. Guštar, T. Anagnos, "Simulation-based Reliability Assessment for Structural Engineers", CRC Press, Boca Raton, USA, 1995.
P. Marek, J. Brozzetti, M. Guštar, P. Tikalsky (Editors), "Probabilistic Assessment of Structures using Monte Carlo Simulation, Background, Exercises and Software", 2nd Ed., ITAM CAS CR, Prague, Czechia, 2003.
Czech Standards Institute, "CSN 73 1401: Design of Steel Structures, Addendum A", CSI, Prague, Czechia, 1998.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description