Regression Modelling of Service Life Prediction of External Wall Painting of Algiers Building

Paint coatings, as a protective layer of external wall building envelopes, are mainly exposed to several degradation agents, including environmental and climate factors. In the case of North African countries, there are no studies about the estimation of the performance of this coating exposed under the conditions of these regions during its service life. For this purpose, this paper presents a statistical study on the evolution of service life (SL) prediction for external wall painting, based on the inspection of the degradation state of 45 building facades located in Algiers, Algeria. The selection of different investigated external wall paintings is mainly focused on considering their age, the distance from the coastal zone, and the facade orientations. The degradation severity of selected coatings was also estimated in this work using the calculation of a numerical index (Sw). Simple regression analysis (graphical method) and multiple linear regression analysis (MLR) are used for the development of a mathematical model. The main results of this work indicate clearly a high correlation between the values predicted by the mathematical model, which corresponds to a high value of coefficient of determination of 0.95. A comparison study is also conducted in this investigation work between the performance of the obtained model and the MLR model developed by previous research about the SL prediction of paint coating. As a result, the obtained model during this study gives good precision despite the moderate quantity of considered external wall painting.