Traditionally, designers of temporary structures for use in construction had little guidance binding on their designs. Some owners, particularly infrastructure operators, provided standards and guidelines that permitted increased allowable stresses for certain temporary conditions. Sometimes the increased allowable stresses were limited to new materials or were subject to other stipulations. However, this practice came from a time when codes were much simpler and, in some respects, more conservative than they are now. Should increased allowable stresses still be used in the design of temporary structures, or is this practice anachronistic?
The answer is not simple and depends on who you ask. Different professional approach temporary works in very different ways. Structural Engineers are typically squeamish about construction means and methods and are often very conservative about soils and other loads commonly supported by temporary structures. Some structural engineers will, incorrectly, claim that a structure has “failed” if the computed factor of safety is below design code values. Geotechnical engineers typically view factors of safety to be a matter of judgment and some deprecate codes and standards and structural design generally. Thus geotechnical engineers will take a more aggressive approach to temporary structures, but may take risks unwittingly, especially when considering elements and systems that are not in contact with soil. Construction engineers are often highly risk-tolerant, but use simple and typically conservative methods for their temporary structures designs. It is hard to find consensus as to design approach, much less a standard of care among such disparate perspectives.
Part of the problem is that while the use of increased allowable stresses in temporary structures is traditional, the logic behind it seems to have been lost to history. The “temporary overstress” is similar in form to the increased allowable stresses for load combinations including wind and, later, seismic effects that were found in legacy codes. These codes permitted allowable stresses to be increased by a percentage, colloquially referred to as an “overstress”, when the structure was subjected to these transitory loads. This practice was intended to account for the fact that the likelihood that multiple types of loads would be at their extreme values at the same time is remote. This concept lives on in the load combinations for modern codes, although no “overstress” is permitted for dead load effects anymore, even when combined with wind and seismic.
However, this logic does not really apply to temporary structures. Temporary structures are typically designed for dead loads, live loads and applicable environmental loads. The loading is similar to those for permanent structures of similar form, but the design life is shorter. I have heard the argument that a shorter design life reduces variability and therefore should reduce the expected extreme value loads. Some argue that loading of a structure is better controlled on a construction site than in an in-service structure. Thus the loads can be better defined and require a smaller safety factor to attain a comparable measure of safety.
The live loads specified in ASCE 7 and model codes for the design on buildings are based on a combination of experience, measurement and statistics. The published values are intended to represent the extreme value within a 50-year service life. A partial safety factor of 1.6 is applied to this with respect to the expected strength of the structure. With respect to the actual or expected service live load, the factor of safety is quite large.
Now consider construction live loads. Sometimes construction design loads are fairly arbitrary, such as a distributed load for a work platform and sometimes they represent a specific event, like a crane load or worker carrying tools. These loads are closer to expected values and have much less built-in conservatism than the occupancy loads used for permanent structures, which are calibrated using extreme value statistics. Therefore, even if these loads are more certain given the access conditions and the short design life of temporary structures, they are more likely to be exceeded than the permanent loads. While you could argue that higher safety factors could be justified for construction live loads, ASCE 37, Design Loads on Structures During Construction, recommends that construction live loads be combined with other loads in a similar manner as specified in ASCE 7 and model codes. ASCE 37 does, however, provide reductions for environmental loads, like wind and earthquake load effects.
Another argument for the use of “overstresses” is that the short design life somehow limits the effects of the loads. Some materials do have a time-dependent response to load. Wood, for example, is stronger when resisting short-term loads than long-term loads and the design codes account for this. The allowable stresses can be increased, typically be 25 percent, when construction loads are being resisted and the specific increase can be adjusted with the expected load duration.
Soil is usually thought of as a load by temporary structure designers, but it is also a material with time-dependent resistance. The load on a temporary earth retaining structure is, in fact, the difference between the load effects in the soil and its internal resistance. The increased, but unpredictable short-term resistance to collapse of freshly excavated soil is the basis for the “stand-up time” concept used in a number of excavation techniques. Soil loads on a retaining structure are expected to increase with time, but there are other effects that appear to reduce soil loads due to soil-structure interaction and geometric assumptions that are not captured in simple analytical and empirical models. Care must be taken to understand why a soil load effect appears to be reduced because some reductions are means of correcting for a conservative assumption in a particular method and may not be applicable to other methods. ASCE7 treats soil loads similarly to live loads so that safety factors are consistent with traditional practice for permanent work, which does not always make physical sense. Therefore, reducing the factor of safety for soil load effects can be justified to some extent provided that the reduction is consistent with the site conditions, the design methodology for the structure and any underlying assumptions.
Steel does not exhibit time-dependent behavior with respect to construction loads, but higher stresses have traditionally been permitted for temporary steel structures by a lot of authorities and practitioners. One justification I have heard for this is the fact that traditional steel design codes did not fully account for the plastic moment strength of beams. This would make sense if designers increased allowable stresses only for braced, compact sections in bending, however, I have seen it applied to other limit states. In addition, modern steel design codes account for plasticity. While properly detailed steel construction is a forgiving material, increased allowable stresses are hard to justify unless there is some excess conservatism in the design.
There are a number of other reasons why the use of increased allowable stresses can be problematic. The material design codes are typically based on tolerances that are tighter than what is practiced in the construction of temporary structures. Temporary structures are often reused or constructed with used material, subjecting them to deterioration and damage. In addition, temporary structures often lack redundancy and are not always properly designed or detailed for stability. These all reduce the actual factor of safety for a temporary structure in ways that are not considered in the design. Finally, the code compliance of a structure designed with increased allowable stresses is questionable. IBC-based codes require that temporary structures be designed in conformance with the code, without specifically differentiating temporary structures for construction and those for other purposes.
So should higher allowable stresses be used for temporary structures in construction? Clearly, it depends on the circumstances. Any temporary structure needs to be designed to perform adequately when subjected to its design loads. There are clearly some differences between temporary and permanent structures, including loads and material responses that are lower in the short term. It is completely reasonable to consider these differences in design. However, temporary structures are still structures and are subject to the same laws of physics and mechanics. The temporary “overstress” used for construction will likely become a thing of the past because it accounts for the right things in the wrong way. Instead, directly accounting for reduced load effects is more rational and more consistent with good design practice.