Probabilistic analysis of degradation of façade claddings using Markov chain models

Silva, Ana; Gaspar, Pedro L.; de Brito, Jorge; Neves, Lu�s C.

Abstract

In this study, the time-dependent stochastic degradation of three types of claddings is analysed. For this purpose, 203 fac¸ades with stone claddings(directly adhered to the substrate), 195 with adhered ceramic claddings and 220 with painted surfaces were analysed. All the fac¸ades are located in Lisbon, Portugal. Their degradation condition was assessed through an extensive field work. Based on the data gathered, Markov chains are used to predict the degradation of claddings and to understand, in some detail, how the characteristics of the claddings contribute to the overall degradation. The results show that the distance from the sea and exposure to damp are significant to the degradation of all types of cladding. The type and size of stone plates also influence the degradation of stone claddings. The exposure to wind-rain action has a high impact on the degradation of ceramic claddings. The models proposed provide useful information on the probability of failure of the claddings; these results are fundamental in the context of insurance policies and in the definition of building maintenance plans.

Citation

Silva, A., Gaspar, P. L., de Brito, J., & Neves, L. C. (2016). Probabilistic analysis of degradation of façade claddings using Markov chain models. Materials and Structures, 49(7), 2871-2892. https://doi.org/10.1617/s11527-015-0692-5

• Journal Article Type: Article
• Acceptance Date: Aug 8, 2015
• Online Publication Date: Aug 15, 2015
• Publication Date: Jul 1, 2016
• Deposit Date: Mar 24, 2016
• Publicly Available Date: Mar 24, 2016
• Journal: Materials and Structures
• Print ISSN: 1359-5997
• Electronic ISSN: 1871-6873
• Publisher: Springer Verlag
• Peer Reviewed: Peer Reviewed
• Volume: 49
• Issue: 7
• Pages: 2871-2892
• DOI: https://doi.org/10.1617/s11527-015-0692-5
• Keywords: Degradation ; Claddings ; Markov chains ; Stochastic analysis
• Public URL: https://nottingham-repository.worktribe.com/output/976277
• Publisher URL: http://link.springer.com/article/10.1617%2Fs11527-015-0692-5
• Additional Information: The final publication is available at Springer via http://dx.doi.org/10.1617/s11527-015-0692-5
• Contract Date: Mar 24, 2016