Condition Monitoring and Predictive Maintenance
Published on : Thursday 06-10-2022
Predictive maintenance normally uses condition monitoring through real-time data obtained with the help of IIoT, says Soundharyaa Nandakumar.
In today’s era, industry leaders are looking for even a tiny improvement to enhance their overall production in this competitive manufacturing environment. Only an experienced plant operator will tell you, your machinery assets are never constant, which affects the production capital .One big method which significantly increases production across multiple manufacturing is the Overall Equipment Effectiveness’ (OEE), which act as major source of performance indicator which measures equipment’s productivity. By implementing the smart factory technologies like Industry 4.0, it is possible to access the information in real time. OEE determines either one of the factors like availability, performance and quality to identify whether the machine is deteriorating, which shows that the machine is in need of maintenance.
This method of predicting maintenance beforehand through OEE has the ability to save manufacturing industry capital with high performance level. One of the many examples are ‘The Worximity Smart Food Factory’, which runs on low-cost hardware system and smart technologies, which can be up and running for hours combined with powerful analytics control panel that provides fast RoI (Return on Investment) to start, which is a roadmap to full IIoT success.
When we talk about maintenance, it should not always be reactive, reacting to the machine condition more or else, reactive maintenance is difficult to plan due to unplanned spontaneous errors, which results in longer downtimes. Risk of errors can be reduced with planned and regular maintenance, compromising equipment run time at once for stipulated short period of time; which is where Condition Monitoring and Predictive Maintenance come in the picture.
The accompanying graphic illustrates how Condition Monitoring and Predictive Maintenance are maintaining the health of manufacturing units with four different questions – ‘What’, ‘When’, ‘Why’ and ‘If’’.
Condition monitoring
Condition monitoring is the process of regular monitoring of machines in terms of speed, volume, temperature and many more. These parameters will serve a physical indicator to determine the machining condition or health. This condition monitoring often offers manufacturing companies the opportunity to know their machines better and react to their conditions more efficiently.
Condition monitoring is often used in combination with predictive maintenance, but also has an effective role in preventive maintenance in order to identify and rectify irregularities in right time to avoid damage. It is normally used in areas where there is mass production and optimisation. In short, condition monitoring ensures reduced downtime and improved reliability in production.
Predictive maintenance
The combination of IoT generating mass data and the analysis of this data by making sense of it with the help of increased use of Machine Learning and Artificial Intelligence, particularly in the industrial sector, is what enables Predictive Maintenance. Predictive maintenance results from analysing massive amounts of data to find where small errors are imminent in the machining process. By this method, minor errors can be rectified before they cause major drawbacks.
Data generated in real time in machine operations by IoT connected smart devices like sensors and related software are generally used to utilise cloud-based analysis to monitor the integrity of machines and prevent catastrophic failures, reducing downtime and raising safety concerns. Predictive maintenance can also offer various consecutive benefits in productivity and efficiency which will explore in greater depth at later period. We also need to understand that there exist challenges in predictive maintenance, but it is worth for future RoI (Return on Investment).
It’s always good to discuss the disadvantages rather than only jotting out the advantages, to get to know about a technology. I mean to say when we get to know technology or a procedure which has been followed for greater success, it is better to know the challenges during and after implementation.
1. Progressive Connectivity via Smart Devices: Connectivity and data communication are the pillars of predictive maintenance. Since the data is
collected via aftermarket smart devices like sensors, the collected data must be transmitted to a central storehouse, where it is monitored and analysed based on insights.
2. Data Security: Information security should be always a major concern for any ‘Connected’ organisation. In this era, any organisation undergoing business with connectivity is more exposed to cyber threats. In case of predictive maintenance, it is very much not guaranteed that outside parties cannot access the performance data.
3. Integration: Integration is the step that assures everything is working fine as it should and implementing predictive maintenance is the process where hardware and software should work along and communicate effectively in real time.
4. Expensive and Time-Consuming: One of the major challenges is the amount of time needed to access and implement a maintenance schedule, and investing in correct maintenance tools and systems will be very expensive, especially when there is business prominence. Right amount of training is needed to cope up with the growing business.
The difference
As organisation adopts condition based or predictive maintenance, it all ends in one single goal of significant RoI for IIoT. The main difference in predictive maintenance and condition monitoring is the timing. Both monitor the health and condition of a rotating asset like a pump, fan, compressor, mixer, agitator, or conveyor. But condition monitoring focuses on real-time conditions, while predictive maintenance has focused on the very early detection of defects 60 or 90 days in advance.
Is predictive maintenance right for industry and why?
For predictive maintenance it is a piece of cake, when it comes to reduced downtime, however with substantial benefits, it still sometimes cannot predict the exact equipment issues as compared with machine learning technology. Nevertheless, any technology or process has its own damages and deterioration, but it drives the future by identifying the shortcoming in early stage, which helps the industries with decision-making, real-time monitoring, which yields good work efficiency.
Generally, any major or minor breakdowns are very unaffordable to any industrial party. It affects their business, process efficiency and widening of their status. Even a small mistake can lead to additional recovering expenses, even production quality and zero transparency within a production process. Predictive maintenance normally uses condition monitoring through real-time data it obtains with the help of Industrial Internet of Things (IIoT), which is hugely considered as a vital factor of Industry 4.0 revolution, which monitors operations dynamically and detects possible future errors and regulate them.
The evolution of fourth industrial revolution (4IR) in providing ideal automation is through smart technologies like Big-data, Internet of Things as mentioned before and machine to machine (M2M) communication methods and correspondingly predictive maintenance actuate on these three main technologies of 4IR and process as one. With a number of benefits by its side like reduced downtime as we all know, adding to it, it is also known for its scalability, equipment monitoring, real time data insights, transparency, and proactive solutions without any fuss.
At last, continuously designing solution-oriented condition monitoring and maintenance systems with grabbing the knowledge needed for the applications with solution implemented in common IoT platforms will greatly increase the OEE of any industry involved.
Soundharyaa Nandakumar has done her Masters in Mechatronics and Cyber-Physical System in Germany and presently working as a Test Engineer, testing Vehicle Electronic Control Units in Connected Drive department. Her previous work experience as an automobile technician has given her a good understanding of how beautiful the car/vehicle is with intricate designs and unbelievable technologies inside. This craze and passion for cars made the decision for her to pursue a career in Automotive Technology.
In her own words, “I always compare cars to women, because we never know what is inside them, how mysterious they are with n number of technologies evolving around every day making them futuristic statements – just like women – undergoing many things in a day to be strong enough to conquer the world.”