Predictive maintenance gets more intelligent with advanced analytics

KONE moves more than a billion people every day. Together with Solita, KONE is enhancing the predictive maintenance capabilities for more than 150,000 elevators with sensor data, machine learning and analytics. The collaboration is part of the ongoing development of an intelligent maintenance service called KONE 24/7 Connected Services.

A global leader in its industry, KONE provides elevators, escalators and automatic building doors as well as solutions for maintenance and modernisation, which add value to the life cycle of any building. The company has more than 1.4 million items of equipment in its maintenance base around the world.

The need to predict and fix faults faster

Traditional elevator maintenance is based on regular maintenance cycles and inspections as well as fault reports from customers. Considering that KONE maintains more than a million elevators globally, the traditional model is both time consuming and expensive.

In 2017, the company launched KONE 24/7 Connected Services. The system enables vast amounts of data from elevator sensors to be monitored, analysed and displayed in real-time. Solita’s expertise was brought in to further optimise the service.

Enhancing predictive maintenance with modern analytics

KONE’s collaboration with Solita has focused on developing cloud and edge analytics to measure the behaviour and condition of the equipment from sensor data. Together, the sensors and elevators form a global Internet of Things (IoT) system transmitting information that can be used to detect faults in real time and analyse symptoms leading to faults before they occur.

With algorithms tailored to different purposes, the data is used to examine the characteristics of an individual elevator. The characteristics relate to the mechanical properties as well as the nature and use of the equipment. By combining the observations, it’s possible to monitor the normal operation of the elevator in real time and to identify possible malfunctions.

In addition, the data is used to gauge significant features related to possible faults in the future. Based on these findings, machine learning models and statistical models that predict faults are designed and implemented. The results of the solution are then translated to inform elevator maintenance.

New capabilities enhance data-driven predictive maintenance

The predictive maintenance system allows KONE’s agile maintenance team to service the equipment faster. With the intelligent cloud-based solutions implemented together with Solita, KONE can better serve their customers, and facilitate safe and reliable people flow all over the world.

“During our collaboration with Solita, we’ve gained deeper understanding and stronger capabilities to produce more sophisticated data analytics solutions. Intelligent and data-driven services are truly at the center of serving our customers now,” says Olli Mali.

“Solita has been involved in building algorithms since the early stages. It has been rewarding to witness the global deployment of the analytics solutions. Over the years, we’ve experimented with numerous options and implemented significant parts of the analytics together,” adds Olli Mali.