Predictive Maintenance solutions and their opportunities

The data floodgates have burst in the Internet of Things – nowadays even coffee machines and car tires are as “smart” as entire production lines. In the old days, a system had to fail before it could be repaired. Now, sensors report imminent defects before they even occur. These examples of predictive maintenance solutions save costs and open up completely new business models for companies.

Predictive maintenance solutions can help companies to save costs and create new business models.

“The hand towels are almost out in washroom 17 on the third floor, there’s no soap in washroom 21 on fourth and toilet paper is running out in 26”. Cleaning staff can use this information to plan their rounds better and more efficiently. Washrooms belong to the most maintenance-intensive rooms in any building. This predictive maintenance solution was developed jointly by the Fraunhofer Institute for Integrated Circuits (Fraunhofer IIS) and full-service provider CWS-boco International GmbH.

New analysis techniques and algorithms process various unstructured data and measurements to provide reliable forecasts. This means devices, machines and vehicles can be maintained before (critical) incidents occur.

No more flat tires thanks to predictive maintenance solutions

This is also true of tire manufacturer Continental who has launched a tire platform for remote monitoring of utility vehicle tire condition. Sensors in the tires continually measure values such as pressure and temperature allowing conclusions to be drawn from the tire conditions before sending these to the platform. If one or more parameters reach critical value, ContiConnect alerts defined recipients who are then able to respond before any acute damage can take place. Other manufacturers such as Goodyear and Pirelli are also currently working on similar predictive maintenance solutions.

Costa Express operates coffee vending machines that need regular refilling, careful cleaning and regular maintenance because fresh milk passes through their nozzles. In this case, the IoT platform provided by Device Insight has improved the availability of the machines and their service. Refilling and maintenance of the machines now takes place precisely so that no employees or customers have to wait for their coffee, espresso or cappuccino and the operator does not miss out on any sales.

The IoT platform as the basis for predictive maintenance solutions

An IoT platform such as CENTERSIGHT NG provides the infrastructure for predictive maintenance solutions and the basis to collect, manage and analyze data streams that supply machines, plants, vehicles and devices. The platform also takes on the role of interface to ERP, production and support systems. It offers all the functions for testing worldwide machine parks and installations from fault detection, repair and notification to data analysis and assessment visualization.

There is a growing trend towards Edge Computing: Data is processed locally by CENTERSIGHT EDGE and analyzed near real-time. If anomalies are found, results are sent immediately to the cloud platform. This reduces data volume and analysis time. Predictive maintenance is only possible if the time factor is exhausted fully. IoT data that is first analyzed when, for example, an elevator gets stuck, offers no added value.

In order to make irregularities or anomalies predictively useful, CENTERSIGHT NG relies, among other things, on Boolean logic. Using the Boolean (named after British mathematician George Boole (1815–1864)) operators AND, OR, NOT or XOR (Exclusive OR), many statements or forecasts can be reached.

For example: IF the temperature of a cold store rises repeatedly above minus 18 degrees Celsius AND power failure can be excluded as cause of error, it must be examined whether the cooling seals have failed, or employees simply forget to close the cold store door.


These simple guidelines can be created with the customer’s domain knowledge and provide a practicable introduction to predictive maintenance solutions. On this basis, algorithms can be further refined by means of machine learning. predictive maintenance and the networking of components and products via IoT platforms have long since allowed mid-sized companies innovative business models, especially in the improvement of services. If machines or installations can send service-relevant information to manufacturers or suppliers on their own and even create tickets when required, the service department is able to react before the customer can even register the defect. A perfect way to increase customer satisfaction.

Recommended posts


Process Optimization with Foresight: Digital Product Passport

Why the digital product passport is already a relevant tool today for making data available along th...

The Asset Administration Shell – a universal tool for data exchange in industry

How the Asset Administration Shell implements the concept of the Digital Twin in industrial practice...

From M2M to IoT & AI: 20 years of moving to a digital industry

The journey from M2M communications to IoT and AI-enabled systems. A look back.