Journal of Structural Engineering

Commonly‐used structural reliability models are reviewed, and the advantages and limitations of each method are discussed with the help of a practical example. A pressure vessel is designed against wind loading. This example is selected because the limit state equation is nonlinear and the wind speed has an extremely skewed distribution. Although the First‐Order Second‐Moment may be sufficient in many problems, the risk involved may still be unknown, even in a relative sense, unless the limit state equation is linear and all the variables are normal. In that sense, the Advanced Second‐Moment method has several advantages. It yields probabilities of failure similar to the variance reduction techniques, at least for the pressure vessel example considered here. If simulation techniques are used to study a problem, some kind of variance reduction technique should be used for economic reasons. For the complicated example considered here, the Conditional Expectation plus Antithetic Variates variance reduction technique is found to be the most efficient. The Monte Carlo Simulation technique is modified to make the simulation more efficient.

The distribution is approximated for extremes of linear combinations of specialized non‐Gaussian models (translation processes) used to represent intermittent loads. These models can be generated from smooth normal processes by nonlinear transformations and have on‐events with a random magnitude of any specified marginal distribution and random shape. The method proposed in this study for load combination analysis is applied to estimate extremes of the combined effect of two exponential load processes.

In most conventional methods of reinforced concrete analysis, cracks are formed in the principal stress direction and are not allowed to change direction with the change in state. This practice leads to crack directions inconsistent with the limit state. There is experimental evidence that the crack directions may change in the course of loading. A simple model of forming cracks in reinforced concrete is proposed. It is assumed that the cracks are formed in the direction of the major principal tensile strain and that the direction can change with the change in strains. The proposed model leads to crack directions which are consistent with the limit state. A numerical algorithm, based on the proposed model, which is suitable for step‐by‐step finite element analysis is also presented. The algorithm is applied to limited available experiments.

This paper reports the experimental investigations that have been carried out to study the ultimate strength and behavior of trapezoidal steel flooring units. The steel deck models were subjected to pure bending loads and the failure was found to occur either due to yielding of compression flange or due to web buckling. Stiffening the compression flange increased the ultimate load significantly. Simple theories using the effective width concept are shown to give reasonably accurate predictions of the failure loads.

The energy absorption in, and the inelastic behavior of, simple structures during strong earthquake excitation are investigated. The purposes of the investigation are to evaluate the performance of simple structures during various types of ground motion, and to attempt to identify more succinctly, than at present, the factors that influence structural deformation and damage. The study focuses on the amount of energy imparted to a structure, the amount of energy dissipated by inelastic deformations and damping, the displacement ductility, and the number of yield excursions and reversals experienced during the excitation. Based on the amount of energy imparted to structures, a possible effective motion criterion is defined.

Prestressed concrete segmental box girder bridges have been used in America for the last 16 yr. The cross section of a segmental prestressed concrete box girder usually has transverse cantilevers, which may incorporate transverse post‐tensioning. The transverse cantilever decks of the Southway Bridge were load tested to verify the design. The experimental results are presented and compared to the design analytical models used to predict the structure's behavior.

The forces induced in braces depend on the magnitude of initial imperfections (lateral bending and twist) and magnitude of applied loads. A method for calculating these forces is presented here. Shear stiffness of attached deck is taken into account. Calculations show that, for beams which derive their lateral stability mainly from the stiffness of the attached deck, additional bracing against rotation or lateral deflection is not able to achieve an important increase in the load capacity of the systems. In those cases, bracings employed merely to straighten the initial imperfections need not be very strong. In cases where decking cannot be counted on to stabilize the roof system, bracings should be designed for stiffness (sufficient to enforce a higher buckling mode) and checked by the methods presented here for required strength. The results are of value to designers of large flat wood roof systems;

Results are summarized of analytical and experimental studies conducted on physical and analytical models of a 15‐story reinforced concrete framecoupled wall structure which was identified as typical of medium‐rise office construction in the western United States and designed in accordance with provisions of the 1973 version of the Uniform Building Code. The primary objectives are to evaluate whether or not: (1) The state‐of‐the‐practice leads to a design that accords with the criteria envisioned by seismic design documents; and (2) the state‐of‐the‐art of analysis of reinforced concrete frame‐wall structures provides reliable analytical assessments of the seismic behavior of structures. The results indicate that designs based on the Uniform Building Code provisions may in several important ways be inadequate and that analytical techniques currently used to evaluate such designs similarly suffer from shortcomings.

Recent failures of breakwaters have been attributed to low impact strength of armor blocks. The simultaneous loss of individual weight and interlocking ability that follows failure of slender units rekindled the interest in cubic blocks. Data generated at the site of Sines, where storms in 1978/1979 caused severe damage to a breakwater, are utilized. The study reports results found in an ongoing program that examines aspects of materials behavior associated with repetitive impact loading of concrete cubes. A simplified version of the continuous damage theory is utilized to study coaxial collisions, leading to determination of parameters such as the speed that causes fracture in a single shock and the damage incurred for collisions at lower impact speeds. Estimates of fatigue‐like effects and accumulated damage on the strength of the cubes are also presented.

This paper describes the extensive Forest Service transportation system, one of the world's largest, and its role in the economy of the United States. About 6,500 of the roughly 13,000 bridges on this 330,000‐mile (531,000‐km) system have wood as a major structural component. Reasons why wood is a viable and competitive structural material are discussed. Inspection procedures to detect decay and conditions conducive to decay are discussed, as well as preservative treatments which are used to minimize decay. A method to determine strength loss due to decay is presented. Timber bridge maintenance is discussed from the standpoint of preventive, early remedial, and major maintenance. Examples of bridge rehabilitation (major maintenance) of deteriorated and substandard timber bridges are shown and discussed.

A theoretical model is presented to evaluate the thermal loading of a concrete roof slab heated by solar radiation. Roof temperatures are computed from standard meteorological data and the material properties of the roof. Then effective temperature is used to characterize in‐plane deformation and thermal moment to characterize bending effects. Comparison with experimental values, recorded in an isolated roof panel for a year, shows the model to represent slab behavior accurately on clear sky days for the mediterranean climate of Adelaide, Australia. Design charts are presented to evaluate the structural parameters for a range of roof constructions in a particular climate. For the computation of characteristic values histograms are presented to show the probability distribution of the structural parameters for specific roofs located in three Australian climate regimes. The frequency distributions are not normal and their form is a function of climate.

The nonlinear finite element technique is used to analyze beam‐to‐column moment connections. In case of flange moment connections, a full scale assembly of the connection has been considered for predicting the load‐deflection behavior, the stress distribution in the panel zones, and the spread of yielded zones at various stages of loading. Connections studied herein are fully welded, flange welded only, and fully bolted moment connections. The analytical results are compared with the results of the full scale tests carried out at Lehigh University. It is shown that a complete connection behavior can be predicted analytically. The web connection considered in this paper is an unsymmetrical one where there is only a beam on one side of the column. A nonlinear stress analysis in the region of the flange connection plate has been carried out by isolating it with suitable freebody. Typical combinations of beams and columns are considered for the study. Through the study of stress distributions in the flange connection plate, the weld required to join the flange connection plate to the column has been proposed.

When perfectly straight beams are subjected to loads without eccentricity, the critical load is the lateral‐torsional buckling load obtained as an eigenvalue for the bifurcation problem of beams. Since a real beam used in practice has not only geometrical imperfections due to initial out‐of‐plane and twisting deformations and eccentricity of loads, but also material imperfections, such as premature yielding because of material nonlinearity or residual stress, the behavior of a real beam is somewhat different from that of a perfect beam. The maximum value of load on a load‐deflection curve of a real beam is called the ultimate load‐carrying capacity or, the ultimate load of a beam. In this paper the lateral‐torsional buckling load and the ultimate load‐carrying capacity is distinguished. Then the relationship between the lateral‐torsional buckling load and the ultimate load‐carrying capacity of beams and the effect of the types of residual stress distribution, as well as the loading conditions and the geometrical imperfection levels on the ultimate load‐carrying capacity are examined.

The force method which so far is used only for the analysis of rigid frames where buckling is disregarded, is here extended So include the influence of the axial compression force on the bending moments. For this purpose the flexibility matrix is properly modified. The relative advantages of the force and deformation methods for a buckling analysis are considered. Tables are included, giving the modified elements of the flexibility matrix together with a systematic design procedure. Numerical examples cover buckling of portal frames, beam columns on elastic support, and stayed columns.