Journal of Structural Engineering

Bridge damage frequencies observed on a 5,500 mile (8,800 km) segment of railroad during a 10‐yr period indicate that the probability of damage is strongly influenced by bridge type and moderately influenced by location and traffic density over the bridge. Sixty‐eight incidents of serious damage were caused by high or shifted loads, fires, floods, collisions and derailments. The risk of fire damage to timber trestles is greater for open‐deck than for ballasteddeck trestles and increases with trestle length. The overall annual probability of serious damage was 0.0016. The annual probability of mechanical damage from derailments was 0.00019. That of high load damage to underpasses was 0.010. For shifted load damage to through trusses, without regard to traffic density, it was 0.030. It was 0.00027 for highwater damage to timber trestles and 0.00036 for fire damage to timber trestles of all lengths. The overall lifetime probability of damage in the order of 0.1, for an assumed 75‐yr life, appears to be an appropriate lifetime probability for serviceability failures.

A systems approach is presented for fatigue reliability analysis of offshore platforms. The probability of fatigue crack occurrence in individual structural joints is considered. A nomograph relating deterministic fatigue life to probability of fatigue cracking is developed based on the lognormal fatigue reliability format. Joints are grouped according to ranges of fatigue life. Fatigue cracking probabilities of joints in each group are determined as a function of service life. Bernoulli trials are used to evaluate the probabilities of 0, 1, 2, 3,… cracks occurring in each group and also in the entire structure. The effects of fatigue life correlation are included. It is shown that the number of joints, years of service life, and correlation effects are important factors in assessing system fatigue reliability. Analysis of a North Sea platform indicates a high level of reliability against fatigue damage.

Direct shear failure in reinforced concrete slabs under impulsive loads is relatively undocumented because of the paucity of data showing failure characteristics. The combined effects of beam action and wave action are likely to be important in developing models to understand the dynamic direct shear phenomenon. The research summarized in this paper makes an initial attempt to understand this phenomenon by considering elastic beam action to describe incipient direct shear failure conditions. The effects of load rate and beam‐end restraint are investigated. Failure curves developed from elastic Timoshenko beam models are compared with experimental data on one‐way slabs which failed in direct shear.

The experimental investigation of the first crack and ultimate load behavior of steel fiber‐reinforced concrete deep beams with openings is reported. The effect of fiber content, positions of openings, and different types of loadings on the behavior of such beams is studied. The experimental results are supplemented and compared with established analytical solutions. They are in good agreement and indicate substantial increase in strength with the increase of fiber content. The significance of the natural load path joining the edge of loading and supporting plates is exemplified through a drastic reduction in strength when it is intercepted by the opening.

An analogy between two systems, each consisting of an elastic beam on elastic foundation, is given. Through this analogy, forces and displacements of one of two beams correspond to the displacements and forces, respectively, of the conjugate one, and field equilibrium and mechanical boundary conditions of one beam correspond to field compatibility equations and kinematical boundary conditions of the other beam. As the beam and foundation stiffnesses of the real system correspond to the foundation and beam compliances of the conjugate one, the problems of statically indeterminate beams not lying on foundation, having uniform or nonuniform cross section can be solved by using simple equilibrium equations on the conjugate beam. The analogy can be applied also when the beam on elastic foundation experiences inelastic deformations both distributed and concentrated. This enables us to solve several practical problems, e.g., temperature change effects, influence lines, secondary moments on statically indeterminate prestressed structures, etc. Applications to cylindrical tanks subjected to rise of temperature and influence lines of nonprismatic shear beams are reported.

A reliability‐based optimization sensitivity analysis technique based on the feasible directions concept is presented for plastic design of redundant structures. This technique is specialized to a formulation of the problem of plastic optimization for minimum weight based on a rational reliability constraint for sizing members. The technique is demonstrated with a structural design example where emphasis is placed on the sensitivity of the reliability‐based optimum solution to both the correlation among strengths and the method for evaluating the overall probability of plastic collapse.

Shear walls play an important role in the wind and seismic resistance of low‐rise wood‐framed buildings. The behavior of shear walls is primarily governed by the nail force‐slip characteristics. The bending stiffness of studs and the shear stiffness of the sheathing play an important role in providing the stiffness. But the studs and the sheathing play a secondary role in defining the load‐deformation properties. New models are presented to represent the shear behavior of these walls. It is shown that the proposed models are comparable to the finite element model, and give results which are in good agreement with the test results. The proposed models are particularly suitable for repetitive applications, e.g., in a nonlinear dynamic analysis.

Uncertainties in structures result from natural hazards, man‐made hazards, variations within common practice, and accepted departures from common practice. However, the major source of structural uncertainty and failures is departure from accepted practice, i.e., human error. The purpose of an error control strategy is to lower the frequency of errors which have large consequences. Sensitivity functions which relate structural errors to structural safety are bases for a control strategy. Sensitivity analysis is performed for several practical cases to demonstrate calculation of sensitivity functions and its importance in establishing an error control strategy.

A new constitutive law for the load resistance versus displacement of equal leg fillet welds is derived from consideration of the mean stresses induced in a fillet weld due to general in‐plane loading of the weld element. Plasticity theory with an isotropic hardening rule is used to derive expressions for the ultimate strength and maximum displacement of the fillet weld. A set of curves representative of the constitutive law is presented. The constitutive law contains the fundamental parameters of the problem, illustrates the anisotropic nature of the response of fillet welds and is presented in invariant, nondimensional form.

Design pressures in circular bins, which have been of concern in recent years, are examined. Shallow bin pressures are distinguished from deep bin pressures. Furthermore, deep bin pressures are examined for funnel flow and mass flow conditions. Pressures due to outside temperature variations are examined. Examples and drawings are provided to clarify the principles described herein.

Provisions of 1982 UBC, ACI 318‐83, and ATC 3‐06 pertaining to seismic shear design of slender walls in mid‐rise construction are evaluated. In the event of major ground shaking in regions of high seismic risk, the actual shear strength demand is expected to equal that associated with the axial‐flexural supply. Thus, the codes' minimum design requirements ought to insure that flexure, and not shear, will control the seismic response during the expected rare, major seismic event in the western U.S. The codes do not implement this condition. Expressions suggested by design documents for computing the shear strength of walls were evaluated by comparing the predicted and measured strengths of 10 wall specimens tested at Berkeley. Although generally conservative, since code expressions do not incorporate the actual shear resisting mechanisms of walls under seismic effects, it is possible for the expressions to mislead the designer to poor shear design. Recommendations are formulated to improve the current shear design procedures by: (1) Relating the shear strength demands to the actual axial‐flexural supply; and (2) incorporating the actual shear resisting mechanisms in predicting shear strength supply of walls.

The deformation properties of concrete significantly influence the behavior of complex concrete structures. These properties are subject to large variabilities and it is therefore not possible to accurately predict the structural response. The sources of these variabilities are first examined, and an example is then presented of the long term deformation calculations in a multiple span bridge. Finally, a procedure using the finite element method, is proposed for predicting the variability in the shortening of the bridge. A number of analyses are performed for selected combinations of material properties, and the deformation at different time steps are calculated. Polinomial models are then fitted to the results at each time step. These models are used to calculate the distribution of the response, the confidence limits and estimates of the decomposition of the variance of the response. The results indicate that the variability in the response can be reduced by performing selected tests on the actual material, and using that data to adjust the numerical material models.

Cold‐formed members enable the engineer to design more cost‐effective transmission towers. Members can be fabricated to closely fit the design requirements. Stiffening lips can increase local‐buckling resistance of thinner sections and ensure members that can be handled without damage during erection. Recommendations for the design of plain and lipped angles produced by cold‐forming are provided. Methods are presented which allow cold‐formed members to be designed by the same criteria presently used for hot‐rolled angles. These methods can also be extended to more complex shapes. Test data provides verification.

The ultimate‐load carrying capacity of relatively wide continuous orthotropic slab bridges has been estimated using the yield‐line theory. Bridges of skew and rectangular planforms under several loading conditions are considered. The influence of the angle of inclination of the interior pier line support on the collapse load is established. The assumed yield‐line patterns, based on a parametric study using a progressive failure analysis, are verified and substantiated by test results from two 2‐span continuous prestressed concrete waffle slab bridges of skewed and rectangular planforms. A method by which a concentrated load is made equivalent to an AASHTO HS 20 truck loading is presented. A design example illustrates the use of the equations derived.

A method is presented for prediction of stresses and strains of partially prestressed concrete structures. Members are assumed to be uncracked under permanent loads, but cracking can occur under transient live load. A superposition procedure is employed to calculate the time‐dependent effects of various loadings including the initial prestress and the changes in stress in concrete and steel due to creep and shrinkage of concrete and relaxation of prestressed reinforcement. The analysis accounts for the presence of non‐prestressed reinforcement and gives the time‐dependent values of axial strain, curvature and stress in concrete, prestressed steel and non‐prestressed steel. Numerical examples are included for calculation of the curvatures in uncracked and cracked partially prestressed beams and the curvature values are employed to determine the deflections.

A simple formula is derived for the moment of inertia of a structural section consisting of three or more areas evenly spaced along a circular perimeter. The result is independent of the axis of orientation and the number of areas. For a finite number of areas, elements of the calculus of finite differences are used to sum the second moments of the areas and differential calculus is used to confirm the limiting case.