United Kingdom

Select your location

United Kingdom


DC/767 United Kingdom, 2016-12-21

ODiN Predictive Maintenance heralds new era in maintenance best practice

Leading drive and control specialist, Bosch Rexroth, has introduced a new system which is set to minimise downtime by bringing greater accuracy to predictive maintenance techniques across a broad spectrum of applications.

ODiN Predictive Maintenance uses machine learning methods to generate knowledge on the state of health of equipment from recorded sensor data, to enable reliable predictions about likely time to failure – a key attribute given the widening adoption of Industry 4.0 techniques.

Instead of the value-based analysis associated with the majority of condition monitoring systems, ODiN employs a model-based approach – for the first time, advancing the process from pure state monitoring to truly predictive analysis and data-driven, anticipatory maintenance.

This is achieved through an initial phase during which a machine learning algorithm determines a normal healthy state for each component based on information from a variety of sensor signals taking readings such as pressure, flow rate, vibration, temperature, and oil quality, depending on the equipment being monitored. This phase may only last a few days if the system in question carries out the same functions under very similar conditions all the time – however, if it is only seldom used, or is used in differing ways to manufacture various products, this phase may be extended.

Following the learning phase, ODiN defines a ‘health index’ for each component being monitored with its data-based model. If an individual measured value temporarily deviates from the tolerance range, an error warning may not necessarily be generated, as wear and tear can seldom be detected from a single signal. However, if the health index deteriorates based on changes in data received from multiple sensors, then the system warns of a problem.

Tapio Torikka of Bosch Rexroth explained: “Diagnosing wear and tear in industrial applications is an extremely complex task. Statistically, there is only a 13 percent probability of an issue being detected by chance, while an expert monitoring the system by traditional means has a 43 percent chance of detecting it. Our system has a detection rate of 99 per cent.

“The system acquires all the necessary information from the sensor data and machine learning methods, then converts this into knowledge. The health index therefore not only shows the state of the assembly currently being monitored, but also gradual changes to upstream and downstream mechanical or hydraulic systems. If movements take longer or require more power, this indicates wear and tear. ODiN gives corresponding instructions in its regular health index reports and helps to create specific recommendations for action.”

He added: “Even ODiN cannot fully eliminate the risk of plant downtime, but we can reduce the risk so significantly that the costs for the system are generally already recouped after the first prevented downtime.”

For further information visit: http://bit.ly/ODiNCondMon


Search press releases


Lynsey Cutler
15 Cromwell Road, St Neots 
PE19 2ES Cambridge
Phone: 01480 223290