Using IoT sensors and AI to predict maintenance and prevent disruption

We worked with IoT technology providers Mindsett, to analyse real-time sensor data to forecast maintenance needs – predicting failures and preventing disruption.

The problem

From broken boilers to faulty deep fat fryers – unexpected asset failures cause disruption, and significant financial losses, to businesses around the world. 

Mindsett, an IoT technology provider, wanted to expand their product offering to their clients in the retail and restaurant industries who suffer from asset failure. Mindsett approached to incorporate real-time machine learning analysis of sensor data into their system to indentify usage patterns and predict asset failure.

The solution

To address the problem, worked with Mindsett to analyse sensor data from stores., alongside the Mindsett technology teams, worked to design a data infrastructure to support the deployment of deep learning models in real-time and at scale across their client base.

We built intelligent maintenance schedules for assets using insights from usage, environment, and service history. The team also created a system to flag unusual measurements based on historic asset performance. The team developed, tested and deployed an AI framework which can correctly predict asset deterioration with over 86% accuracy.

Correctly predicts asset deterioration

The impact

  • Intelligent Maintenance Schedules created bespoke maintenance schedules for assets taking usage, environmental factors and service history into account

  • Predictive Asset Failure

    AI was used to predict asset health - identifying appliances at risk of developing faults and pro-actively intervening to prevent disruption

  • Real Time Monitoring and Alerts

    The system produced real time energy and asset usage alerts based on intelligent thresholds learned from historic performance.

If you’re interested to find out how AI technology could transform your business email or use the contact form below.