KRESNIK: A top-down, statistical approach to understand the fire performance of building facades using standard test data

Abstract

The facade is one of the most complex parts of a building, performing multiple objectives of value to the occupants. The frequency of facade fires in tall buildings is increasing, therefore it is crucial to understand the behaviour of such facades in a fire, but there is currently no theory, model, or series of experiments that allows this understanding. This paper takes a top-down, data driven approach to understanding facade behaviour by analysing a unique database, named KRESNIK, containing 252 commercial facade fire tests, the first time such data has been analysed. We found that the outputs from these tests were correlated, which could be used to gain more information of facade performance than simply pass or fail; and that the different layers of a facade can have a significant effect, particularly the addition of a cavity. Rainscreen facades performed the worst (45% failed), whereas none of the ETICS or sandwich panels in KRESNIK failed. We also found that the choice of cladding material of these rainscreens is the most important factor in driving their fire performance, but that neither its total fuel nor its conductive resistance can predict fire performance. Finally, we found that repeated tests of identical facades could have major variations in the outputs, but that whether the facade ignited or not tended to remain consistent across repeats. These results help to identify critical factors in facade flammability, better informing engineering decisions, and contributing to the design of safe tall buildings.