Smart Roofing Decisions: An AI-Based Recommender System Integrated into RoofNav

Abstract

AI is being applied more and more in the construction industry to aid in decision-making, specifically material selection for buildings and optimizing roofing systems. In this paper, an innovative AI-based recommender applied in RoofNav is proposed for smart roof decision-making by merging environmental data analytics and machine learning. The system integrates various data, such as FM Approved roofing assemblies, climate data, and user behavior, to suggest personalized and context-sensitive recommendations. The proposed methodology uses a hybrid machine learning-based recommendation model which combines collaborative filtering and content-based filtering to enhance the accuracy and relevance of selected suitable roofing material for various kinds of buildings in different locations. The RoofNav integration delivers a user-friendly experience, so building professionals are able to quickly make informed decisions that are in compliance with local codes and performance requirements. This serves as an example for how AI in construction is proving to be a game-changer, encouraging sustainability, safety and cost-efficiency in the creation and instillation of roofing systems. The ensemble model of combining ensemble learning, collaborative filtering via SVD, and a meta-learner delivered the best results in Precision, Recall and MRR. It was 38% faster and more user-friendly than previous if-and-only-if systems, and 90+% of users “enthusiastically” favored its design. This further illustrates how it helps guide homeowners new and old through complicated roofing choices with confidence.