Predictive Modelling of Building Airtightness for Life CycleBased Assessment of Energy Performance

Abstract

This paper presents a data-driven approach for assessing building airtightness and infiltration-related energy effects using a predictive mathematical model based on artificial neural networks and preliminary evaluation against blower door measurements. The model enables preliminary estimation of airtightness parameters (e.g., n50) during early design stages, supporting building performance assessment without relying exclusively on costly on-site testing. The research further conceptually integrates predicted infiltration-related energy effects into Life Cycle Assessment (LCA) and Life Cycle Costing (LCC) frameworks, supporting sustainable decision-making. Results indicate that predictive approaches may support energy performance evaluation and contribute to improved building envelope design in early design stages.