Metodologia para estimar a vida útil de elementos construtivos, baseados no método dos fatores

Abstract

This Work, based on parts of the ISO 15686, presents and discuss about a deterministic method, Factor Method (FM), that permits to predict material service life and their components and planning their maintenance and replacement necessary periodicity. The precision of the estimates of this method was analyzed in this Work, and an alternative equation to evaluate service life was developed, using multiple linear regression (MLR), derived from FM pertaining to ISO 15686-1 and especially the Factor Method. The MLR is broadly used to evaluate products that interact with independents variables. Finally, this alternative equation was used to compare its estimate random results with the FM random ones. After evaluating the examples on chapter 4, the Author noticed that the Factor Method is very important to estimate service life of materials and components and the alternative equation too. Two case studies were done in this Work, in the first case the total service life of three collapsed structures were estimated. First a concrete structure similar to a building that was located in the Metropolitan Area of Recife (Brazil) and then the total service life of an balcony from a building located at the city of Rio de Janeiro (Brazil) using both the Factor Method and the alternative equation, and finally the total service life of the basement and superstructure of specific buildings called box from the Metropolitan Region of Recife using the Delphi Method together with the Factor Method was realized too. In the second case, the service life estimation of a softwood window was done using the FM and the alternative equation, other estimations of service life were done the same way by using the alternative equation and the FM. Despite the equations presented in this Work have been developed specifically to estimate service life of materials and components, the estimates showed that the precision of the total service life of the first case approximated to the real case studies, proving the potentiality of the methods