How AI is changing predictive maintenance | Jargon buster
A growing band of suppliers is adopting AI and machine learning to improve the age-old challenge of spotting problems in buildings before they arise, writes Chintan Soni.
In one sentence, what is predictive maintenance?
Predictive maintenance simply means digitally identifying performance and health issues ie early signs of degradation causing inefficiencies and leading to failure, leveraging machine learning and AI technology to achieve this.
What are the benefits of predictive maintenance?
Underperforming plant and equipment in buildings and factories can use excessive energy to start with and ultimately lead to failure if they are not maintained on time. So being alerted to these issues through smart technology such as ours can not only save on energy costs, but it can also reduce how much energy is needed, cutting your carbon footprint in the process. Predictive maintenance also reduces downtime of the building as issues are resolved before the repair costs become expensive for owners and occupiers.
How does a building generate the data it needs for predictive maintenance?
AI and machine learning algorithms process data from existing IoT sensors which are embedded across equipment. These IoT devices would typically measure flow rates, energy and utility consumption, vibration of plant, and temperature. Some of these data points may already be available within a building management system, but owners and occupiers may need to install additional IoT sensors to enrich their insights.
How much tech expertise is required from the real estate team?
Translating hard data into understandable, actionable insights which can make a positive difference to the performance of a building and its carbon footprint is the key to success. Our goal, and that of other tech-led changemakers in the space, is to roll-out a solution that existing engineering teams can adopt seamlessly.
Is there a risk from cyberattacks or data privacy incidents?
Security is a high priority for commercial and industrial real estate owners, and emerging sustainability technology solutions need to have world class systems and infrastructure, such as AWS, to ensure operations remain safe. In addition, technology should not read-write to the BMS without human intervention.
Chintan Soni is CEO and co-founder of Ecolibrium, a machine learning sustainability and decarbonisation platform for commercial and industrial real estate, which recently expanded into the UK from India
Also in this series