This chapter presents and discusses the principles, methods and the associated limitations that currently are seen as the state-of-the-art in structural safety. The basis for understanding the design philosophy of modern design codes is provided. Innovative concepts in safety, starting with definitions of risk, reliability, fragility and a new definition of structural robustness are presented. Uncertainties are discussed and a risk management framework for structural design is proposed. A probabilistic structural design philosophy is presented detailing a new methodology for analysing structural fragility and the robustness of structures against failure. An example is presented determining the robustness of a falsework structure against collapse. Strategies to enhance structural robustness and structural safety are given. An improved design methodology for temporary structures is presented and detailed, and an example is provided. Finally, the chapter discusses the use of reduction factors when determining design action values for the design of temporary structures.
Top5.1 Introduction
The design of engineering structures can essentially be defined as a continuous process of making difficult engineering decisions based on the available knowledge and under the severe constraints imposed by society and nature. Any structure can be analysed in an integrated system made of exposures, hazard events and consequences. However, no matter the existing time interval, budget size and analysis capacity it is not possible to determine precisely the behaviour of any structure due to uncertainties. The key element for structural safety is the impact of uncertainties in the available knowledge.
In the traditional approach, engineers resort to structural design codes to make decisions. These documents are developed specifically to address areas where significant past experience exists and where critical societal risks are not involved. Thereby, design codes are established for the purpose of providing a general, simple, safe and economically efficient basis for the design of ordinary structures under normal loading, operational and environmental conditions. Design codes not only greatly facilitate the daily work of structural engineers but also provide the vehicle to ensure a certain standardization within the structural engineering profession which in the end provides a uniformity of reliability of structural performance and enhances an efficient use of the resources of society for the benefit of the individual. In countries such as the USA, engineers are required by law to adhere to design codes, whereas in Europe, engineers are allowed to produce designs which deviate from the design code if they can demonstrate that the design fulfils the reliability levels specified in the design code. The latter may ease engineers in the process of producing innovative engineering solutions.
However, problems do exist. Current design codes are normally based on semi-probabilistic limit states design, such as the Limit State Design (LSD) methodology. However, design codes were calibrated to provide an adequate reliability only at the individual element level. Therefore, resistance safety checks are merely considered at a local level (e.g. a cross-section or an individual element) and the designer has insufficient control over the analysis and selection of preferred mode, or modes, of failure of the designed structure with respect to critical enabling/triggering hazard events. As a result, the global behaviour is not directly accounted for and the design efficiency and the global target reliability may not be achieved in practice.
As highlighted by Starossek (2006), the safety of the structure depends not only on the safety of all the elements against local failure but also of the system response to local failure. The implied assumption that the adequate resistance of the structure is guaranteed by the resistance of its elements is generally not valid, see Starossek & Wolff (2005). In addition, Ellingwood (2008) pointed out that
(...) no attempt was made to rationalise the calibrated reliabilities in explicit risk terms; thus, they are related to social expectations of performance only to the extent that reliability benchmarks obtained from member calibration to historical practice can be related to such expectations.
Current code design philosophies may also limit the tools at disposal of the designer to optimise the structure to specific performance objectives. To do so would require the use of different partial factors for each component type, size, structural arrangement, type of loading, type of usage, etc. (CIRIA, 1977), determined based on a risk informed decision-making process. As a consequence, the reliability target levels used during calibration of design codes are in practice the best estimates of the actual values. They are often called notional target reliabilities.
It can be concluded that the present basis for design does not assure optimal design in terms of resources allocation and risk acceptance. As a result, the traditional standards-based approach is becoming increasingly inadequate to handle the allocation of limited resources for structures design, operation, repair or improvement, in a climate of growing public scrutiny.