Smart Maintenance: 9 Key Concepts You Need to Know
Date: 2018-11-1 11:00:30
We are in the midst of a transformation that started already a few years ago, nicknamed as the ‘4th industrial revolution’. According to Klaus Schwab of the World Economic Forum it’s all about ‘industrial convergence’. That is the merger between the physical, digital and biological world. This is possible thanks to developments such as:
• 3D printing,
• Virtual reality and
• Internet of Things
When we focus the ‘4th industrial revolution’ on maintenance, we hear terms such as:
• Predictive Maintenance,
• IIoT and
• Edge Computing
• Digital Twin
Can you still keep track? The following nine terms update your basic knowledge of Maintenance 4.0.
1. Preventive Maintenance
Preventive maintenance acts on the principle of ‘prevention is better than cure’. Instead of waiting for a malfunction to occur, the intelligent software schedules a maintenance plan. The goal is to prevent failures before they occur. This contrasts with the old approach of ‘Run to failure’ maintenance which is reactive.
2. Industrial Internet of Things
IIoT is one of the basic blocks of the 4th industrial revolution. Engineers are linking more and more components, installations and objects. This enables new analysis and insights. The most elaborated application today is predictive maintenance combined with big data analysis.
3. Predictive Maintenance
Predictive maintenance goes a step further then replacing a certain part after a fixed number of running hours. The intelligent software looks at the part’s health based on ‘condition monitoring’. Condition monitoring uses data from:
• Vibration measurements,
• Oil analysis,
• Or infrared measurements.
The intelligent software predicts if a failure is likely within a certain time frame. Engineers can thus schedule and deliver maintenance better and decrease maintenance costs.
4. Big Data Analytics
Monitoring your key installations delivers a large amount of data. This ‘Big Data’ contains a wealth of information. It is possible to predict the ‘unpredictable’ when you link information streams from inside and outside the company.
The PwC report ‘Mainnovation’ about Predictive Maintenance 4.0 shows this becoming reality. Top companies provide continuous asset monitoring with warnings based on predictive techniques. Regression analysis is one of those techniques.
5. Cloud Computing versus Edge Computing
The evolution to store more and more data and to perform calculations in the cloud continues. Both for individuals and companies. Yet there are good reasons in the industry for doing Edge Computing. With Edge Computing your data remains close to its source for processing. It is a method to optimize cloud computing by performing data processing near the source of the data. It is the edge of the network. This is faster, safer and cheaper when you split between data stored and processed locally, and data sent to the cloud.
6. Artificial Intelligence
Large internet companies pump billions in research and development in the field of artificial intelligence. New breakthroughs follow each other faster and faster. The industry is working with cobots. They are robots that collaborate with and learn from human colleagues. Inspection drones and cleaning robots also start playing a larger role in maintenance.
7. Predictive Analytics
Predictive analytics encompasses a variety of statistical techniques from predictive modelling, machine learning, and data mining that analyze current and historical facts to make predictions about future or otherwise unknown events.
8. Prognostic Maintenance
Prognostics is an engineering discipline focused on predicting the time at which a system or a component will no longer perform its intended function. This form of maintenance builds on predictive analytics and maintenance. It uses machine learning, pattern recognition, and other advanced techniques like ‘neural networks’ and ‘neural fuzzy systems’.
9. Prescriptive Maintenance and Analytics
The most advanced option in maintenance. Prescriptive maintenance tries to answer the question: ‘What should we do to achieve X?’. It’s based on:
• Big data,
•Complex event processing,
Prescriptive goes a step further than predictive maintenance because it not only reflects the possible results of a particular approach but also evaluates which approach is the fastest or most efficient.