SPS Drives 2017 Predictive maintenance & smart monitoring

Predictive maintenance is one of the most frequent industrial use scenarios of the Internet of Things (IoT). Because companies save money whenever they avoid a machine outage or production breakdown. Deutsche Telekom's smart monitoring uses artificial intelligence to recognize anomalies in machines.

Machine sensors and log files produce huge volumes of data which, if analyzed properly, can be transformed into valuable information. Smart monitoring recognizes relevant data patterns and gives machine experts the right tools to identify and analyze irregularities and anomalies in real time all around the globe. If a deviation exceeds a given threshold the system activates an alarm. It supports users in four ways:

As an early warning system
The system triggers a preventive alert before machine outages occurs.

Deep analysis
Thanks to a range of KPIs (Key Process Indicators) and interactive visualization, smart monitoring can get to the bottom of what caused an anomaly.

Root cause analysis
Even where damage has already occurred, smart monitoring helps to analyze the root cause. The system compares incidents with previous anomalies to identify the cause(s) of damage.

Risk assessment
Smart monitoring allows unlimited numbers of machines to be assessed simultaneously. This gives companies a detailed picture of their entire machinery. They can see, for example, what machines are at greatest risk of failure and thus need to be replaced soon.

The benefits of smart monitoring at a glance:

• The system recognizes machine anomalies automatically - without any expert knowledge.
• Machine outages and the associated costs can be avoided. 
• Artificial intelligence autonomously filters and selects the key information from the huge volume of machine data.
• Web-based dashboards provide a convenient overview and are intuitive to use. 
• Smart monitoring is a robust procedure and the ideal starting point in predictive maintenance.

Three partners – one solution

"Predictive maintenance", the solution presented at SPS IPC Drives 2017, is the result of three partners working together: Eaton as industrial automation specialist integrates machines into IoT environments. In this way Eaton's XV 300 and XC 152 controls collect local machine data and control the requisite processes. Machine control data are securely transferred to the cloud via an OPC-UA compliant message protocol. T-Systems, the second partner, provides the necessary cloud platform, the predictive maintenance application and corresponding condition monitoring and advanced analytics features, while Deutsche Telekom, the third partner, ensures connectivity.

Thanks to this predictive maintenance solution, companies can set threshold values for practically any machine data. The system then alerts the users if this threshold is exceeded, allowing them to quickly recognize where a machine is at critical risk and take countermeasures. Diagrams and graphs provide both up-to-date and historical data in a clear way.

Certified data centers 

T-Systems provides all predictive maintenance services from the cloud. Customers therefore do not have to invest in their own data centers and software infrastructure. This is particularly advantageous where companies experience major fluctuations in the amount of computing capacity they require, meaning these resources are not always put to full use. All T-Systems cloud services for predictive maintenance are based in certified highly-secure data centers that comply with strict data protection and privacy provisions.