Abstract

The rehabilitation of aging educational buildings has become increasingly important in the context of sustainable campus development and adaptive reuse of existing infrastructure. This study proposes an integrated BIM-based framework for the rehabilitation of underutilized academic spaces through the combined application of Building Information Modeling (BIM), biophilic interior design principles, and Artificial Intelligence (AI) predictive modeling. The methodology was implemented in a case study involving non-functional areas within the Faculty of Aerospace Engineering at the National University of Science and Technology POLITEHNICA Bucharest. Autodesk Revit was employed to develop a parametric digital model of the existing structure, support spatial reconfiguration, and assess environmental and functional performance indicators throughout the rehabilitation process. To evaluate the effectiveness of the proposed framework, multiple performance criteria were considered, including spatial efficiency, daylight performance, material sustainability, acoustic quality, and user-perceived visual comfort. Furthermore, a synthetic dataset generated through parametric simulation was utilized to train and compare four machine learning algorithms (Multiple Linear Regression, Support Vector Regression, Random Forest, and Artificial Neural Networks) to predict user comfort based on spatial and environmental variables. The rehabilitation strategy resulted in an 18% increase in usable floor area, a 26% improvement in average daylight factor, a 25% increase in renewable material utilization, and a 38% reduction in estimated acoustic reverberation time. Simultaneously, the predictive modeling revealed that the Artificial Neural Network (ANN) provided the highest accuracy (R2 = 0.91) in capturing the non-linear relationship between biophilic design elements and perceived interior quality. By integrating Gilbreth’s principles of cognitive ergonomics, the AI framework actively prevents the rigid, purely quantitative optimization associated with “Digital Taylorism, The findings demonstrate that the proposed BIM-integrated rehabilitation framework can support both technical optimization and user-centered environmental enhancement in educational facilities. The study contributes a transferable digital methodology for sustainable academic building transformation, combining geometric precision, predictive environmental performance assessment, and human-centered design principles within a unified rehabilitation workflow.

IPC Classification

Keywords