To main content

jAInitor

The jAInitor project aims to develop new solutions that help building managers detect faults in HVAC systems earlier and fix them faster. Instead of relying on threshold alarms or manual inspections, jAInitor will use artificial intelligence (AI) to monitor ventilation systems and provide clear, understandable alerts, giving the building what can be described as a digital janitor.

Contact persons

Buildings account for a significant share of energy use in Norway, and in commercial buildings, nearly 40% of electricity use goes to ventilation, heating, and cooling. Research shows that faults in these technical systems lead to poor indoor climate, higher costs, and unnecessary energy use equivalent to 0.5–1 TWh per year. 

Illustrasjon: ChatGPT

jAInitor will:

1) Build a knowledge base of faults in HVAC systems. The goal is to make it easier to detect faults, understand their consequences, and handle them more effectively.

2) Develop flexible and scalable fault detection and diagnosis (FDD) tools. By structuring and storing data in a systematic way, the project will remove technical barriers and make such solutions easier to adopt.

3) Increase the use of FDD tools among building operators by identifying barriers, building trust, and ensuring that solutions are adapted to practical operational needs.

4) Reduce the time required to identify and resolve faults in HVAC systems by streamlining workflows and decision-making processes.

Key facts

Partners

The jAInitor project is led by SINTEF Community, with the Norwegian University of Life Sciences (NMBU) as a research partner, and industry partners including KLP Eiendom, Fram Eiendom, Coop Eiendom, Mustad Eiendom, Agilitek, Coor, EM Systemer, Enoco m/Alti Forvaltning, Properate, VirtualHouse, Piscada, Aase Teknikk, Soundsensing, Toma Eiendomsdrift og Autility.

In addition, international experts from Aalborg University, Penn State University, and Lawrence Berkeley National Laboratory (LBNL) will contribute to the project.

Funding

Financed by the Research Council of Norway (project no. 358549) and Norwegian industry partners. 

Project duration

2025 - 2029