Developments in International Bridge Engineering

In order to resolve the durability problem of expansion joints and bearings, the integral abutment bridge (IAB) has become more and more popular. For integral abutment bridge, choosing a suitable pile foundation type is a challenging problem, because the substructure is fixed with superstructure to bear the load together. In this paper, the design of the pile foundation in a fully integral abutment bridge (FIAB) in China was analyzed. A finite element model was built by the commercial software MIDAS considering soil-structure interaction and construction stage simulation. A sensitive analysis was carried out to investigate the influence of different pile foundation types on the mechanic performance of the IAB. The results show that when the circular pile is used, the stress of pile, negative moment and tensile stress of girder are smaller than those when the rectangular pile is used. With the increase of pile diameter, the stress and displacement of pile decrease, while the bending moment of pile and the negative moment and tensile stress of girder increase. For rectangular piles, with the increase of cross-sectional length-width ratio, the bending moment of pile, negative moment and tensile stress of girder decrease; while the stress and displacement of pile increase.

In this paper we study the response, under live loads, of two new cable stayed bridges (Baluarte and Carrizo), recently open to traffic. Both bridges are located in the northern pacific mountains of Mexico along the Mazatlan-Durango new highway at approximately 50 km from the coast. These bridges happen to be the most important bridge structures of the highway. Both bridges have the same fan cable pattern and similar type of superstructure. However the response of both decks under live loads is quite different as will be shown. This paper describes also the numerical analyses to evaluate the bridges structural response under static and moving loads mainly. From this evaluation and assessment, conclusions and recommendations related to their structural stability are provided. The activities described were performed before opening the bridges and comprised also the instrumentation with fiber optic sensors, live load tests and ambient vibration tests.

The current provisions for the braking force in the Eurocodes are much more demanding than previously enforced national codes from most European countries. Hence, safety assessments of existing bridges may show a lack of compliance with the new safety requirements of current maintenance codes, making the relevant authorities responsible for their strengthening. The current load model for the braking force was derived from a deterministic evaluation of traffic configurations with characteristics that do not correspond with actual traffic measurements. This procedure differs from models describing vertical traffic loads, where the effects of traffic were analysed within a probabilistic framework in compliance with the partial safety factor design method. This paper investigates the advantages, limitations and hypotheses of a probabilistic model for the braking force when compared with deterministic models. Original results for the characteristic value of the braking force on road bridges are presented, which consider a realistic time-history of bridge crossing vehicles generated by a traffic microsimulation tool with input data from a Swiss Weigh-In-Motion station. The results are compared with the braking force of the Eurocodes, highlighting the role of the probability of braking and the influence of the dynamic characteristics of bridges on the characteristic value of the braking force.

In 2009, the Federal Highway Administration (FHWA) published the “Load Rating Guidance and Examples for Bolted and Riveted Gusset Plates in Truss Bridges”. According to this document, main truss member gusset plates were required to be examined for five limit states based on either the Load Factor Rating (LFR) or the Load and Resistance Factor Rating (LRFR) approaches. However, further investigations showed that this methodology generally resulted in conservative gusset plate ratings. The National Cooperative Highway Research Program (NCHRP) was sponsored to address these concerns by investigating the shear, tensile and compressive strength resistance of gusset plates at the strength limit state. The findings of this program resulted in a set of propositions for revisions to the American Association of State Highway and Transportation Officials (AASHTO) Manual for Bridge Evaluation to be implemented in 2014. It is observed that these new provisions for the LFR and the LRFR methods result in significant variation in rating results for the different limit states. In this paper, a comparative study is completed on a gusset plate from the Charles M. Braga Jr. Memorial Bridge, which is located in Fall River/Somerset, MA, U.S.A., to investigate the dissimilarities between the load rating results obtained based on both the 2009 and the 2014 provisions for the LFR approach. The primary aim of this work is to provide a better understanding on the proposed changes for the gusset plate load rating approach described in the Manual for Bridge Evaluation, 2nd Edition, 2014 Interim Revisions.

Examination into FEA techniques for modelling the performance of a specially designed hybrid multi-girder bridge deck system with three model configurations of varying complexities showed good correlation with experimental global load-deflection behavior as well as local strain behavior at the base of the bridge system. For ascertaining peak load conditions, the simplest of the three model configurations was found to provide the best accuracy of results where the peak condition was estimated within a 2–12 % accuracy range for the FEMs studied. However, the local behavior in the transverse composite system, particularly in the horizontal GFRP rods, obtained from FEA greatly underestimated (3.5–9 times) the tensile strain experienced, even when tube-in-tube elements were implemented to represent the unbonded movement of GFRP rod through the surrounding plastic ducts. Good representation of overall behavior in the transverse composite system, however, was obtained with tube-to-tube elements, with transverse curvature exhibited. Interestingly, the addition of tube-to-tube elements did not affect the overall global performance exhibited through FEA, resulting in identical load-deflection curves with a comparable FEM without the use of tube-to-tube elements. The computational time required to complete the model with tube-to-tube elements was also less than that required for the model without tube-to-tube elements. Further study is recommended to determine and eliminate the unknown constraint placed on the GFRP rods, to allow for greater accuracy in the prediction of tensile strain.

The current study aims to determine the dynamic vibration characteristics and structural earthquake performance of the approach viaducts of Bosphorus Suspension Bridge. Ortaköy viaduct at European continent and Beylerbeyi viaduct at Asian continent have a length of 231 and 235 m, and five spans and four spans, respectively. The viaducts are side span of the bridge and are supported at the base instead of hanger elements. Based on the project specifications of the viaducts, 3-D detailed finite element model of the viaducts is developed by the use of frame elements, and natural frequencies and corresponding mode shapes of the viaducts are presented in the study. Paying attention to these considerations, pushover (POA) and nonlinear time-history analysis (NTHA) are performed to attain structural earthquake performance of the viaducts according to Turkish code for the earthquake design of railways bridges (TSC-R/2008) and Caltrans (CALTRANS-2001) seismic design of steel bridge. Accordingly, the earthquake performance of the bridge is presented for each non-linear analysis method considering these codes.

Solar radiation developing non-linear temperature distribution at the depth of the concrete structures may cause significant stress changes along the long span box type bridges. Segmental concrete box bridges are one of the example in which stresses change not only due to annual uniform temperature changes but also due to daily temperature changes that can develop a second type of thermal difference; namely, thermal gradient, which forces to have internal thermal stresses within a cross-section. Over the years it has been observed that nonlinear temperature distribution develop at the depth of box girder type bridges can cause bending moments as high as the ones generated due to live loads that may result in concrete cracking. American Highway Association of Transportation Officials (AASHTO) bridge specifications recognize the use of thermal gradient loads in design since 1989. In this specification, the U.S. is divided into 4 zones per the country’s solar radiation zones and gives some gradient values to be applied through the depth of the girder. The aim of this study is to construct a simple similar solar radiation map for Turkey to be used in design of segmental bridges. Temperature and solar radiation data from eight cities in different regions of Turkey are collected, and used in analysis of a box girder bridge model through a thermal finite element analysis program. Thermal differences through the depth of the girder are determined to obtain the design gradient values for the bridge that can be hypothetically constructed in different parts of Turkey. The end results are used to develop a recommendation for a simple solar map that is related to thermal gradients to be used in design of segmental bridges.

Risk assessment and management has gained more and more popularity in bridge engineering due to its direction on preventing or alleviating unfavorable situations during the life cycle of bridges. In the current applications, risk management is recognized as an extension of risk assessment to provide management strategies. In fact, risk management is essential to the improvement of decision efficiency and validity since it is the direct procedure connecting risk assessment and on-site users. To enhance the influence of risk management, management oriented risk assessment is proposed. Introductions to the basic conceptions of risk assessment and management are presented. Detailed instructions on the proposed approach are provided with consideration to different management requirements and stages. Four applications of the management oriented assessment are introduced in detail on how the approach is realized in practical situations.

The main bridge of Nhat Tan Project is a six span continuous cable stayed bridge under construction in Hanoi, Vietnam. On completion, the 1500 m bridge will be the longest cable stayed bridge in Vietnam, with 4 main spans and 2 side spans of lengths 300 and 150 m respectively. The bridge will have 8 traffic lanes and has a total deck width of 35.6 m. The superstructure of the bridge consists of a composite girder and New Parallel Wire Strands (New PWS) stay cables with a total of over 14,500 metric tons of steel, 37,500 metric tons of reinforced concrete deck slab and 1800 (220 Nos.) metric tons of prefabricated stay cables. This paper details the limitations faced in development of a construction method to ensure the safe, efficient and expeditious construction of the superstructure. In addition, it describes the innovations made to overcome challenges that arose during the implementation of the construction method.

The economic performance of concrete bridges can be improved in certain design situations by using the balanced lift method, a new bridge construction method developed at the Vienna University of Technology. This method proposes to build the bridge girders in a vertical position and to rotate them into the final horizontal position with the aid of compression struts. The compression struts decrease the span of the bridge and allow lighter bridge girders, compared to bridges built by the balanced cantilever method or incremental launching, to be obtained. The balanced lift method can be applied for bridges with high and low piers enabling the construction of bridges over deep or low valleys. The first application of the new method is the construction of two bridges with low piers for the Austrian road management company (ASFINAG) crossing the rivers Lafnitz and Lahnbach on the S7 motorway in the south-east of Austria.

The assessment of the aerodynamic performances of the Third Bosporus Bridge (BB3) has been realized through tests in two different Wind Tunnels, CSTB and Politecnico di Milano (POLIMI) and with different scale factors. The design process of a super long span bridge is strongly influenced by the wind actions on the bridge itself and the definitions of wind loads and wind induced dynamics cannot be done without relevant experimental campaigns. The definition of the local wind characteristics is the preliminary experimental test to be done, and the final verification of the bridge response to turbulent wind is the last stage of the wind design. For the Third Bosporus Bridge all these activities have been undertaken in close collaboration by the two cited laboratories. The collaboration granted a cross check of the results and of the test methodologies and hence assured a good reliability of the collected experimental data-base. In particular, at CSTB two sectional models with different scale factors 1:100 and 1:25 have been tested to assess the bridge stability, the response of the deck to vortex induced vibrations and to define the wind profile on the different lanes, and consequently the wind lateral loads on passing vehicles. Moreover, at CSTB a model of a tower has been tested in a very large scale and high Reynolds number to understand the wind interaction of the stand-alone tower during the erection stages. At POLIMI the overall wind response have been tested using a full bridge 1:180 aeroelastic model to define the bridge response to turbulent wind and to check the bridge stability limit also during the erections stages. Finally a large 1:50 scale multi modal aeroelastic model simulating the torsional and vertical bending deck behavior was also tested in order to check in a larger scale possible vortex shedding induced vibrations as well as for a cross check with CSTB of possible Reynolds Number effects on the porous wind screens. The results of the experimental activities gave to the design team all the information needed to consider the wind response and highlighted the very good behavior of the bridge under wind actions from all the point of view: stability, vortex induced vibrations, wind loads and effects on the passing vehicles.

This paper makes use of different ground motion selection and scaling methods for the identification of predominantly first-mode engineering demand parameters (EDPs) of bridges under earthquake excitation. Two groups of ground motions are selected for this purpose. The first group, expected to result primarily in the first-mode response, is selected using the conditional mean spectrum (CMS) method. The second group, which serves as the reference for comparison, is selected to match a chosen scenario response spectrum. Since the shape of the chosen scenario spectrum allows higher mode response, the ground motions in this second group are considered as the ground motions with higher mode effects. Both groups of ground motions are selected using a method that seeks to match the mean and variance of the target spectrum. Comparison of the nonlinear time-history analysis responses from the two groups for three chosen ground motion scenarios indicates that higher mode effects are more pronounced on column displacements and deck accelerations than the column shear forces.

This study proposes an innovative restraining system for the improvement of the earthquake resistance of precast concrete I-beam bridges mainly in the longitudinal direction. The proposed method is based on the connection of the continuous deck slab of the bridge through the sidewalks with both abutments and the elimination of the end expansion joints. In this way, the bridge obtains four restraints (one per wing-wall), which behave as tension ties during the deck contraction and have the ability to reduce the seismic movements of the deck. The resultant in-service requirements due to creep, shrinkage and thermal effects are properly accommodated. Each sidewalk is divided to zones that are anchored to the deck and zones that have the ability to slip on the deck. The efficiency of the proposed method is investigated by utilizing a precast I-beam bridge of Egnatia Odos Motorway in Greece. This bridge has been converted to a ductile bridge system by using active seismic stoppers at the head of the piers.

Oblique incidence waves play a vital role in the spatial variation of seismic ground motions. It is important to consider the oblique incidence of seismic wave in a valley or in other complex topographies. The numerical simulation method of free field with uniform topographic under oblique incidence waves is studied first. The accuracy is then tested by analytical solutions. A finite element model of bridge-soil system is built to analyze the seismic response of a continuous rigid-frame bridge with high piers located in a deep valley with the consideration of oblique incidence waves. The topographic effect on amplitude and distribution of internal forces where determined for different oblique incident angles and different shear wave velocities of the soil. The dynamic response of this bridge under P-wave and SV-wave are summarized. The results show that the angle of the oblique incidence waves has greater effect on internal forces of the piers and piles in deep valley than in uniform field shapes. If the bridge is located in a valley, the effect of oblique incidence waves on the bridge should be considered.

Bridge condition rating is evaluated based on the material condition of the secondary and primary bridge components. This paper aims to investigate the influence of the condition of the primary bridge components to the overall condition of pre-stressed concrete beam bridge (PCBB), reinforced concrete beam bridge (RCBB) and steel beam bridge (SBB). Four primary bridge components namely surfacing, deck slab, beam/girder and abutments are used as input parameters and bridge condition rating as output parameters. This study utilizes multiple linear regression analysis (MRA) and artificial neural networks (ANN) to investigate the variance of the bridge condition rating with respect to the condition of the primary bridge components. The MRA results show that 62.83, 91.77 and 86.18 % of the proportion of the variance in the condition rating of PCBB, RCBB and SSB are explained by all the primary bridge components in the range of the training data set. Meanwhile ANN yields 67.35, 90.54 and 81.77 % for PCBB, RCBB and SSB, respectively. The results indicate that the condition rating of surfacing, deck slab, beam/girder and abutments highly contribute to the condition rating of RCBB and SSB, however for the PCBB, the influence is slightly lower. In term of modeling, MRA shows better performance for RCBB and SBB; however ANN seems suitable for PCBB.

Bridges are prone to bearing deterioration due to aging, environmental causes and/or traffic overloading, and, as a result, a bridge originally designed as simply supported may be behaving differently due to changes in support conditions. The latter can lead to damage, not only to the bearings, but also to the rest of the structure due to unforeseen variations in the distribution of internal forces. In this paper, static load testing is used to evaluate the true behavior of the boundary conditions of an existing bridge and how they affect the overall structural response. The subject of investigation is a 16 m span concrete bridge subjected to the forces applied by a truck loaded with quarry aggregate. Static measurements are obtained for four positions of the truck and used to tune a mathematical model of the bridge until achieving an accurate resemblance with the experimental data. Boundary conditions are idealized with linear rotationary springs, which allow considering all possible degrees of rotational restraint at both end supports. It is shown how those values of the spring constants that best fit the measurements for a single loading case can be misleading and differ significantly from the optimal solution found combining all loading cases together.

Seismic design procedures are in transition from strength-based to performance-based approaches. Faster and larger computers and improved nonlinear behavior models have paved the way. Nevertheless, validations of these predictive models have typically been confined to experiments conducted at the component scale. Controlled experiments at the system level are scarce. At the present time, the next best option for model validation is to use dynamic data collected from instrumented structures. These data include recorded accelerations due to impact excitations, ambient forces, weak or distant earthquakes, and strong ground motions. Different techniques are needed to extract useful information for such diverse types of data. Herein, we examine a long-span bridge located in California-which is instrumented through the joint efforts of Caltrans and the California Strong Motion Instrumentation Program (CSMIP)-as a case study for demonstrating how and what types of information may be extracted from recorded data that bear consequences for engineering practice. Some of the highlights of this study include the observation of some inconsistencies between data and meta-data, calculation of wave delays, and response prediction with updated finite element models.

Ikitsuki bridge is a three span continuous steel truss bridge completed in July, 1991. Approximately 18 years after the completion, a crack was found in a diagonal member in the vicinity of an intermediate support. In order to identify its cause, fracture surface of the crack was observed and a long-term monitoring of wind and vibration of some diagonal members has been carried out. The study reveals that (1) the crack was initiated and propagated as fatigue crack; (2) vibration of the target members is induced by the wind of 6–8 m/s and more than 15 m/s in the direction approximately normal to the bridge longitudinal axis; (3) the members mainly vibrate in the plane of truss and the vibration frequencies coincide their 1st natural frequencies; and (4) the maximum stress range induced by the vibration due to 6–8 m/s wind is approximately 30–40 N/mm², while that due to wind more than 15 m/s can be 195 N/mm². These wind-induced vibrations are thought to be the main cause of the fatigue crack.

Different from bridges on land, bridges under marine environment are subjected to tide and wave. In order to understand the actual performance of bridge under marine environment, a structural health monitoring system (SHMS) is designed and implemented on a curved continuous steel box girder bridge in the Hangzhou Bay of China. Through the implementation of the SHMS, both environmental parameters and structural response are monitored, including wind, temperature, vibration acceleration, and bearing deformation. By analyzing the monitoring data, characteristics of the environment and structural response are obtained, including the wind field characteristics, the temperature distribution of the steel box girder and the structural dynamic characteristics. From the monitoring results of the girder vibration acceleration, there is an obvious vibration phenomenon found in the lateral direction. Further studies show that the structural vibration has a direct relationship with the tides in the Hangzhou Bay. The obvious vibration is induced by regular ebb and flow, because the lateral modal frequency is as low as about 0.5 Hz which is in the range of the tidal frequencies. Moreover, the foundation scour caused by tide will lower the structural integral stiffness and then the natural frequencies, which may make matters worse. Meanwhile, finite element method is used for structural characteristics analysis and structural response analysis. Comparing the theoretical calculation results and the measured ones, the structural finite element model is verified and modified. And the modified model is used in evaluating and predicting the safety and status of the structure. In the end, some conclusions and management suggestions are given for the bridge under extreme conditions.