Structural Utilisation and Windows

In much of the work about embodied energy minimisation, we explore the structural utilisation of members in a building. This is usually defined as the ratio between an actual performance value and the maximum allowable performance value which is deemed limiting for a structural member. It allows us to explore how close to the limit state (be that serviceability or ultimate) a member is.

Median values for structural utilisation in a building can then be determined to allow us to compare and make general conclusions. A brilliant example is found in:

Moynihan, M. and J. Allwood, Utilization of structural steel in buildings. Proc. R. Soc. A, 2014. 470(20140170).

In this post I raise two questions for further analysis:

1) How can design choice be captured in a utilisation ratio?

Utilisation ratios are calculated with an underlying assumption of sensible choices of structural form. As an example, a floor beam bent about its minor axis may exhibit a utilisation ratio of 1.00. However, simply rotating the beam by 90º to bend about its major axis would reduce the elastic utilisation by about 90%. Understanding how sensible choices of structural form are made in the design stage is therefore a key part of ensuring materially efficient designs. So, the question for our work is - how can we incorporate the need for sensible choices of structural form into the calculation of a utilisation ratio? Or perhaps there is an alternative mechanism to measure this.

2) Can we define a "structural safety window"

Utilisation ratios give us a measure of how close we are to the design performance of a member. So if the effects of a design action are equal to the design resistance, the utilisation ratio is 1.00.

However, this does not mean the member is physically failing (in the case of the ultimate limit state) as we have within our "design" values numerous partial safety factors.

In medicine, particularly for opioids, a "therapeutic window" is defined. This is the range between the median effective dose (ED50) - the dose that has the desired effect in half the population - and the median lethal dose (LD50). As the ratio LD50/ED50 widens, the safer the drug is from overdose rise.

In structural design, we might find an analogy that allows an illustration of how far from actual collapse (in the case of ULS) a member is. This is effectively "hidden" from designers at present in the use of reliability indices and subsequent partial factors. Defining an equivalent ratio between median likely loading and median collapse capacity might help to illustrate the difference between these values.