The paper concerns reliability analysis of steel trusses under fire conditions with taking into account randomness of buckling coefficient, modulus of elasticity's and yield strength's reduction factors. To assess the reliability system analysis was employed. The calculations of appropriate standard deviations of random variables were realized in Mathematica program. Two types of trusses were analysed: statically determinate and indeterminate, so respectively serial and mixed system were used. The results, gotten for different levels of coefficients of variation of reduction factors were presented.

The paper focuses on the system reliability of steel trusses with correlated variables. The correlation between bearing capacities of bars was considered. Two static truss schemes were considered. Nodal forces were the only load. The Finite Element Method analysis was conducted in Robot Structural Analysis program. To conduct system reliability analysis it is essential to find cut-sets, it was realized by stiffness matrix spectral analysis. Then reliability analysis was performed in Sysrel module of Strurel computing environment. First Order Reliability Method was used as the base, Subset Simulation method was used to check the correctness of the results. The sensitivity analysis of reliability index enabled the authors to draw conclusions, which variables have the greatest influence on the reliability of the structure. The effects of actions and bearing capacities were assumed to be the only random variables and that the excessing the bearing capacities of bars is the only way the structure can get into failure area.

In line with the principles of modern design a building structure should not only be safe but also optimized. In deterministic optimization, the uncertainties of the structures are not explicitly taken into account. Traditionally, uncertainties of the structural system (i.e. material parameters, loads, dimensions of the cross-sections) are considered by means of partial safety factors specified in design codes. Worth noticing, that optimal structures are sensitive to randomness design parameters and deterministic optimal solutions may lead to reduced reliability levels. It therefore seems natural to extend the formulation of deterministic optimization with the random scatter of parameter values. Such a formulation is offered by robust optimization and reliability-based design optimization. The applicability ofRBDOis strongly dependent on the availability of the joint probability density function.Aformulation of non-deterministic optimization that better adapts to the design realities is robust optimization. Unlike RBDO optimization, this formulation does not require estimation of failure probabilities. In the paper using the examples of steel beams, the authors compare the strengths and weaknesses of both formulations.

The objective of the article involves presenting innovative approach to the assessment of structural reliability analysis. The primary research method was the First Order Reliability Method (FORM). The Hasofer–Lind reliability index in conjunction with transformation method in the FORM was adopted as the reliability measure. The implicit limit state functions were used in the analysis. The formulation of the random variables functions were created in the Matlab software by means of neural networks (NNs). The reliability analysis was conducted in Comrel module of Strurel computing environment. In the proposed approach, Hybrid FORM method (HF) used the concept in which NNs replaced the polynomial limit state functions obtained from FEM (Finite Elements Method) for chosen limit parameters of structure work. The module Comrel referenced Matlab files containing limit state functions. In the reliability analysis of structure, uncertain and uncorrelated parameters, such us base wind speed, characteristic snow load, elasticity modulus for steel and yield point steel are represented by random variables. The criterion of structural failure was expressed by four limit state functions – two related to the ultimate limit state and two related to the serviceability limit state. Using module Comrel values of the reliability index with the FORM method were determined. Additionally, the sensitivity of the reliability index to random variables and graph of partial safety factors were described. Replacing the FEM program by NNs significantly reduces the time needed to solve the task. Moreover, it enables the parallel formulation of many limit functions without extending the computation time.