Next in the series – today I will talk about “Predictive Maintenance”

Predictive Maintenance is one of the most popular use case of IIoT / Industry 4.0 and probably also one of the most misunderstood. As the word ‘predictive’ suggests IIoT system will ‘predict’ a possible fault and period in which it is likely to occur. Based on this ‘prediction’, maintenance can be done before the asset fails. But how does the system do this prediction?

Popular belief in IoT world is that you need big data and AI/ML for predictive maintenance and yes…this is one of the methods to ‘predict’ maintenance. But you can predict need for maintenance via ‘Condition Monitoring’ as well. Some people term this as ‘Prescriptive Maintenance’. This method relies on expert knowledge and IIoT systems convert this knowledge into rule base that acts like an expert monitoring the asset 24×7. Unlike ML based method, this method yields results right from day 1 as it doesn’t need historical data to ‘learn’ asset condition.

There are many misunderstandings about Predictive Maintenance using ML. Here is a blog that outlines these