The effects of AI based predictive maintenance from life cycle perspective

Project title: The effects of AI based predictive maintenance from life cycle perspective
Project leader: Shoaib Azizi, Sustainability Assessment & Managements, SEED, and Farzin Golzar, Heat & Power Technology, ITM
Participating companies/organisations: KTH
Financed by: Digital Futures
Project duration: 2026-2028

This research project develops and validates an integrated framework that combines AI-based predictive maintenance with Life Cycle Assessment (LCA) for smart buildings. Using the KTH Live-In Lab as a testbed, the study analyzes real-time data from more than 150 sensors collecting information every 10 minutes. The project aims to create machine learning models that predict failures in HVAC, piping, and other technical systems, helping extend component lifespan, reduce maintenance costs, and improve operational efficiency. The framework also links predictive maintenance with dynamic LCA to assess environmental and economic impacts, including trade-offs related to sensor deployment and digital infrastructure. By evaluating both the benefits and environmental burdens of ICT systems throughout their lifecycle, the project addresses an important research gap in sustainable building digitalization. The interdisciplinary collaboration integrates expertise in building engineering, environmental science, and digitalization, supported by advanced simulation tools and KTH’s computing infrastructure.

Contact