This thesis presents an approach to evaluate and develop a live load model. Although the main purpose of this work is with the impact of truck traffic on bridges, the data presented in this work can be used in many engineering fields that are concerned with truck characteristics of geometry and weight. Data from two different WIM stations were considered. One in Fernão Dias highway in the State of São Paulo which is comprised of a same-direction two adjacent lanes and the sample is comprised of 20 months (September 2015 to August 2017). The second station is in Rio Grande do Sul State. This road is a same-direction three adjacent lanes. The sample is comprised of 78 days (March 2014 to June 2014) In order to evaluate and develop a new live load model, an approach to compute load effects in terms of bending moments and shear forces is proposed. It makes use of single and multiple truck presence to evaluate the live load effects for different bridge spans. Three cases of multiple presence are considered: following, side-by-side and staggered. The proposed approach to evaluate the multiple truck presence effects is compared with the approach used by AASHTO LRFD. The approach for estimating the bias factors shows that considering only full correlated trucks is too conservative, mostly for short spans where there is a lack of occurrences, especially following events. On the other hand, taking into account no correlation at all yields very low bias factors. At last, a more rational live load model was developed based on WIM data. Another purpose of this thesis is to use existing Brazilian bridges to calibrate the live load model as in NBR7188:2013. Reliability analysis is performed with sixty existing Brazilian bridges. The bridges are taken from different states of Brazil. Out of the sixty bridges, 39 are prestressed and 21 reinforced concrete bridges. Those bridges are located in five different states: Pernambuco, Ceará, Bahia, São Paulo, Minas Gerais. Probability of failure was estimated in terms of moment and shear for interior girders and box girders. Only ultimate state limit was considered. It was found that reliability indices are higher in prestressed bridges when compared to reinforced bridges. Also, the reliability indices tend to decrease as the span length increases. This means that for larger spans the probability of failure is higher than the ones for shorter spans.
[1]
Geoffrey L. Kulak,et al.
Fatigue tests of riveted bridge girders
,
1995
.
[2]
Yang H. Huang,et al.
Pavement Analysis and Design
,
1997
.
[3]
A study of brazilian concrete strength ( non-) compliance and its effects on reliability of short columns 1
,
.
[4]
Tommy H.T. Chan,et al.
Bridge live load models from WIM data
,
2002
.
[5]
Michel Ghosn,et al.
Reliability Calibration of Bridge Design Code
,
1986
.
[6]
Michel Ghosn,et al.
Protocols for Collecting and Using Traffic Data in Bridge Design
,
2008
.
[7]
Michael P. Collins,et al.
COMPRESSION RESPONSE OF CRACKED REINFORCED CONCRETE
,
1993
.
[8]
Andrzej S. Nowak,et al.
Reliability of Structures
,
2000
.
[9]
Andrzej S. Nowak,et al.
CALIBRATION OF LRFD BRIDGE DESIGN CODE
,
1999
.
[10]
Juan R. Casas,et al.
A comprehensive traffic load model for bridge safety checking
,
1997
.
[11]
Geoffrey L. Kulak,et al.
Fatigue strength of two steel details
,
1982
.
[12]
Túlio Nogueira Bittencourt,et al.
Single and Multiple Presence Statistics for Bridge Live Load Based on Weigh-in-Motion Data
,
2017
.
[13]
F. Vecchio,et al.
THE MODIFIED COMPRESSION FIELD THEORY FOR REINFORCED CONCRETE ELEMENTS SUBJECTED TO SHEAR
,
1986
.