TY  - CONF
  - O'Connor A; Pakrashi V
  - 11th International Conference on Applications of Statistics and Probability in Civil Engineering
  - Recent Advances in Reliability based Assessment of Highway Bridges
  - 2011
  - Unknown
  - Validated
  - 1
  - ()
  - Often in bridge assessment the failure of a deterministic assessment at some limit state is takento imply that the structure requires either rehabilitation or replacement. There is however a third option, probability-based safety assessment, which is often ignored by bridge owners/managers but which if considered can lead to considerable savings by avoiding the need for costly repair or rehabilitation. The principles of probabilistic assessment are no different than the techniques used in code derivation. A minimum safety index is specified for the structure, dependent upon its importance, expected failure mode etc. and at each limit state the computed level of safety of the structure is required to exceed this minimum. Therefore probabilistic assessment can be  related to the derivation of a bridge specific code for the structure. The approach benefits from a lack of generalisation inherent in design codes without compromising the level of safety of the structure. A logical progression from probabilistic assessment of individual structures is the concept of safety based maintenance management. The basic premise lies in acceptance of the concept that when it becomes impossible or too costly to maintain bridges to their as-built standard, bridge owners must at least ensure that the bridges fulfil specified requirements of structural safety. Safety based bridge maintenance management is attractive as it enables bridge owners/managers to extend service lifetimes by reducing or postponing costly major rehabilitation projects. Instead focused repairs are performed on specific critical components of the structure to ensure the overall level of safety remains above the minimum requirement. The critical structural components are determined from deterministic assessment. Significantly, at no stage in the process is the required level of safety compromised. Central to the process of probability-based safety assessment is accurate modelling of the resistance and load. Original structural drawings and material records may be employed to provide information on the distributions of the material variables and consequently the parameters governing structural resistance. Bayesian updating of these distributions may be performed using results of bridge inspections, testing etc. Modelling of the loads to which the structure is subjected is often performed using codified assessment models or site surveys of loading. It is critical to assess random variables as accurately as possible to reduce the level of uncertainty. Of all loads to be modeled, by far the most variable are live loads, traditionally prescribed based upon subjective assumptions of the maximum expected axle, vehicle or combination of vehicles. These assumptions led to conservative and/or inconsistent design load effects and consequently,inconsistency in safety levels among structures. Using of Weigh in Motion (WIM) data in probabilistic safety assessment reduces uncertainty in the load-modelling process. Statistical parameters governing traffic flow at site may be accurately recorded and modelled and realistic estimates of load effects can be obtained through realistic simulations of multiple transverse and longitudinal presences. The response of the structure is also modelled with greater accuracy as the WIM system enables computation of the actual influence surface for the structure. Therefore the actual structural behaviour, incorporating any damage or change of structural system is modelled rather than that as idealised from finite element models or theory. It is howeverimportant to recognise that just as probability-based assessment attempts to model structural resistance degradation with time, some effort is required to allow for evolution of traffic with time. This paper presents through example the benefit of some of the aforementioned recent advances for reliability based safety assessment and maintenance management of highway bridges.
DA  - 2011/NaN
ER  - 

@inproceedings{V96224204,
   = {O'Connor A and  Pakrashi V},
   = {11th International Conference on Applications of Statistics and Probability in Civil Engineering},
   = {{Recent Advances in Reliability based Assessment of Highway Bridges}},
   = {2011},
   = {Unknown},
   = {Validated},
   = {1},
   = {()},
   = {{Often in bridge assessment the failure of a deterministic assessment at some limit state is takento imply that the structure requires either rehabilitation or replacement. There is however a third option, probability-based safety assessment, which is often ignored by bridge owners/managers but which if considered can lead to considerable savings by avoiding the need for costly repair or rehabilitation. The principles of probabilistic assessment are no different than the techniques used in code derivation. A minimum safety index is specified for the structure, dependent upon its importance, expected failure mode etc. and at each limit state the computed level of safety of the structure is required to exceed this minimum. Therefore probabilistic assessment can be  related to the derivation of a bridge specific code for the structure. The approach benefits from a lack of generalisation inherent in design codes without compromising the level of safety of the structure. A logical progression from probabilistic assessment of individual structures is the concept of safety based maintenance management. The basic premise lies in acceptance of the concept that when it becomes impossible or too costly to maintain bridges to their as-built standard, bridge owners must at least ensure that the bridges fulfil specified requirements of structural safety. Safety based bridge maintenance management is attractive as it enables bridge owners/managers to extend service lifetimes by reducing or postponing costly major rehabilitation projects. Instead focused repairs are performed on specific critical components of the structure to ensure the overall level of safety remains above the minimum requirement. The critical structural components are determined from deterministic assessment. Significantly, at no stage in the process is the required level of safety compromised. Central to the process of probability-based safety assessment is accurate modelling of the resistance and load. Original structural drawings and material records may be employed to provide information on the distributions of the material variables and consequently the parameters governing structural resistance. Bayesian updating of these distributions may be performed using results of bridge inspections, testing etc. Modelling of the loads to which the structure is subjected is often performed using codified assessment models or site surveys of loading. It is critical to assess random variables as accurately as possible to reduce the level of uncertainty. Of all loads to be modeled, by far the most variable are live loads, traditionally prescribed based upon subjective assumptions of the maximum expected axle, vehicle or combination of vehicles. These assumptions led to conservative and/or inconsistent design load effects and consequently,inconsistency in safety levels among structures. Using of Weigh in Motion (WIM) data in probabilistic safety assessment reduces uncertainty in the load-modelling process. Statistical parameters governing traffic flow at site may be accurately recorded and modelled and realistic estimates of load effects can be obtained through realistic simulations of multiple transverse and longitudinal presences. The response of the structure is also modelled with greater accuracy as the WIM system enables computation of the actual influence surface for the structure. Therefore the actual structural behaviour, incorporating any damage or change of structural system is modelled rather than that as idealised from finite element models or theory. It is howeverimportant to recognise that just as probability-based assessment attempts to model structural resistance degradation with time, some effort is required to allow for evolution of traffic with time. This paper presents through example the benefit of some of the aforementioned recent advances for reliability based safety assessment and maintenance management of highway bridges.}},
  source = {IRIS}
}