Industry 4.0 and Predictive Maintenance
26 May 2017
Predictive maintenance is a key product of Industry 4.0 as it allows businesses to anticipate downtime, efficiently resolve issues and make effective business decisions. Joshua Dugdale, technical manger at the MTA (Manufacturing Technologies Association) explains more
The fourth industrial revolution has different considerations depending on the sector in which you work or the type of business you run. For some, it will be about making smart machines or products, for others it is about creating a smart factory and for some it will be about having visibility of their supply chain. However, despite these different applications, you can argue that they are all based on the same three fundamental ideas: digital machines and products, interconnection and the analysis of big data obtained from the monitoring of the machines or products.
Making machines digital requires a manufacturer to integrate a variety of sensors into the design. These sensors can include the measurement of pressure, temperature, speed, torque, power, vibration, flow rate, and load to name but a few. The intention is that these sensors continuously provide data on the condition and state of the mechanical, electrical and fluid systems and components within a machine, which are subject to wear and tear, and consequently require maintenance throughout service life.
An ever-increasing number of companies want their machines to be interconnected and to have the option of remote running, freeing people from being in front of a machine while it is running and allowing them to work on other things. This means there will be an ever-increasing trend for sensors to be placed permanently within a machine. Consequently, this will increase the level of development for software and applications which can process the big data being obtained from the sensors, which in turn, opens up the possibility of being able to monitor and know the condition of a machine at any given point in time. This will allow issues to be identified before they occur, and for predictive maintenance to be undertaken. Predictive maintenance is a key product of Industry 4.0 as it allows for businesses to anticipate downtime, efficiently resolve issues and make good effective business decisions.
Maintenance has traditionally been either corrective (i.e. repair of a broken component) or preventative (i.e. replacing a component in anticipation of its failure). Predictive maintenance however is based on detecting patterns of failure within data and consequently being able to change parts when the sensor data shows they are about to fail. Predictive maintenance will involve capturing current data and comparing it with historical data so as to predict the future life of a component and consequently when it might fail.
Predictive maintenance therefore is only as good as the historical data sets that you are able to compare the current data against. This could mean that large initial investment is required to setup a system where you can apply predictive maintenance and bring benefit to the company. However, despite the high initial costs, the advantages of predictive maintenance are many and provide an immediate benefit to a company. With the condition of a machine or product being constantly monitored the benefits are:
- Causes of unknown failures are identified before they occur and are consequently rectified.
- Serious breakdowns and minor faults leading to unplanned downtime are avoided – they are resolved before they become an issue.
- Maintenance leading to downtime is accurately planned into the business schedule.
- The service life of all components is increased as parts are only changed when required – decreasing the cost spent on replacement components.
- Spare parts stocks are minimised, with their future purchase being accurately scheduled.
- Increased understanding of the reliability and productivity of machines.
The fourth industrial revolution is upon us and as part of that there will be an ever-increasing number of sensors on machines to make them digital. This will allow for machines to be remotely run and monitored and for an increased understanding of their condition, at any given point in time, to be attained. This understanding will allow for predictive maintenance to be performed as data from those sensors will reveal when components are going to fail. In conclusion, predictive maintenance will bring a variety of benefits to a company, from helping them reduce the amount of time their machines are down and not creating income, to helping them improve the decisions they make for the business, allowing them to be more competitive, more productive, and more profitable.