1987Deep Blue (University of Michigan)Requires access

System Reliability Models for Highway Bridge Analysis.

Jiyuan Zhou

Open publisher page 4 citations

Abstract

Recently, there has been extensive research into bridge reliability analysis, especially in the area of reliability-based design and evaluation. In previous research, bridge reliability analysis has been based on component reliability. A bridge, however, is a structural system, with resistance dependent on redundancies, loading, load sharing, configuration and position of the truck. Due to these complexities, the overall reliability of a bridge system may significantly differ from the component reliabilities for girders, slabs, connections, etc. This study deals with the problem of determining the reliability of a bridge system. The study develops an integration method to estimate the statistical parameters for bridge load and resistance. The method is similar to Monte Carlo simulation, except that it approximates statistics of resistance or load effect from relatively few structural analyses. The procedure is convenient to implement, and allows for the direct use of available bridge analysis programs. Based on the integration method, the resistance models for bridge girders and systems were developed. The mean-to-nominal ratios and coefficients of variation of the moment capacity have been obtained for composite steel girders, reinforced concrete girders, and prestressed concrete girders with different sectional dimensions and geometries. The statistics of a bridge system were calculated by using the integration method and girder resistance data. The reliability of a bridge system was then obtained by combining the statistical parameters of system resistance and system load effect in the first-order second-moment formula. The procedure has been established and demonstrated on a steel girder bridge. The reliability indices were also calculated for various bridge girders. The system reliability was compared with the girder reliability for a simple span girder bridge. The comparison shows that the bridge system reliability is much higher than the bridge girder reliability. This is due to load sharing and redundancies of a bridge system. The effect of correlation between capacities of the girders on the system reliability was also investigated. The analysis indicates that the higher the correlation, the lower system reliability. It suggests that a bridge system behaves like a parallel system.

About this research paper

What this paper is about

Recently, there has been extensive research into bridge reliability analysis, especially in the area of reliability-based design and evaluation. In previous research, bridge reliability analysis has been based on component reliability. A bridge, however, is a structural system, with resistance dependent on redundancies, loading, load sharing, configuration and position of the truck. Due to these complexities, the overall reliability of a bridge system may significantly differ from the component reliabilities for girders, slabs, connections, etc. This study deals with the problem of determining the reliability of a bridge system. The study develops an integration method to estimate the statistical parameters for bridge load and resistance. The method is similar to Monte Carlo simulation, except that it approximates statistics of resistance or load effect from relatively few structural analyses. The procedure is convenient to implement, and allows for the direct use of available bridge analysis programs. Based on the integration method, the resistance models for bridge girders and systems were developed. The mean-to-nominal ratios and coefficients of variation of the moment capacity have been obtained for composite steel girders, reinforced concrete girders, and prestressed concrete girders with different sectional dimensions and geometries. The statistics of a bridge system were calculated by using the integration method and girder resistance data. The reliability of a bridge system was then obtained by combining the statistical parameters of system resistance and system load effect in the first-order second-moment formula. The procedure has been established and demonstrated on a steel girder bridge. The reliability indices were also calculated for various bridge girders. The system reliability was compared with the girder reliability for a simple span girder bridge. The comparison shows that the bridge system reliability is much higher than the bridge girder reliability. This is due to load sharing and redundancies of a bridge system. The effect of correlation between capacities of the girders on the system reliability was also investigated. The analysis indicates that the higher the correlation, the lower system reliability. It suggests that a bridge system behaves like a parallel system.

Why it matters

OpenAlex reports 4 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Recently, there has been extensive research into bridge reliability analysis, especially in the area of reliability-based design and evaluation. In previous research, bridge reliability analysis has been based on component reliability. A bridge, however, is a structural system, with resistance dependent on redundancies, loading, load sharing, configuration and position of the truck. Due to these complexities, the overall reliability of a bridge system may significantly differ from the component reliabilities for girders, slabs, connections, etc. This study deals with the problem of determining the reliability of a bridge system. The study develops an integration method to estimate the statistical parameters for bridge load and resistance. The method is similar to Monte Carlo simulation, except that it approximates statistics of resistance or load effect from relatively few structural analyses. The procedure is convenient to implement, and allows for the direct use of available bridge analysis programs. Based on the integration method, the resistance models for bridge girders and systems were developed. The mean-to-nominal ratios and coefficients of variation of the moment capacity have been obtained for composite steel girders, reinforced concrete girders, and prestressed concrete girders with different sectional dimensions and geometries. The statistics of a bridge system were calculated by using the integration method and girder resistance data. The reliability of a bridge system was then obtained by combining the statistical parameters of system resistance and system load effect in the first-order second-moment formula. The procedure has been established and demonstrated on a steel girder bridge. The reliability indices were also calculated for various bridge girders. The system reliability was compared with the girder reliability for a simple span girder bridge. The comparison shows that the bridge system reliability is much higher than the bridge girder reliability. This is due to load sharing and redundancies of a bridge system. The effect of correlation between capacities of the girders on the system reliability was also investigated. The analysis indicates that the higher the correlation, the lower system reliability. It suggests that a bridge system behaves like a parallel system.

Key concepts: Bridge (graph theory), Reliability (semiconductor), Computer science, Transport engineering, Engineering, Civil engineering, Forensic engineering, Reliability engineering

Related papers

Back to paper searchBrowse research topicsOriginal source
System Reliability Models for Highway Bridge Analysis. — Research Paper | ScholarLens