Predictive vs preventive maintenance — what’s best?

In the case of industrial machinery, emergency repairs are known for extending periods of unnecessary downtime. Both predictive and preventive maintenance are helpful to keep industrial equipment in good condition and to tackle problems before they occur. However, there is a big difference between the two.

Manufacturers constantly adjust to cope with rising competition. Consumer trends rapidly change, and pressure to deliver high-quality products quickly. Equipment downtime can seriously compromise a business' bottom lines

Manufacturers experience an average of 27 hours of downtime per month because of equipment failure, according to, maintenance specialist Senseye. This results in yearly revenue losses in the multi-millions.

Past vs present

Preventive — or preventative — and predictive maintenance are often used interchangeably. Usually, in reference to maintenance strategies that allow manufacturers to act before equipment fails. Both methods are vastly superior to reactive maintenance.

Reactive Maintenance

This is where equipment runs until emergency repairs are needed. Thus, it often costs companies four to five times as much as proactive maintenance options. Operations and Maintenance Practices Guide, Version 3.0 confirms this.

However, preventive and predictive are not the same.

What is the difference between preventive maintenance and predictive maintenance?

What is preventive maintenance?

Preventive maintenance involves checks at regular intervals, regardless of the equipment’s condition. It relies on best practise guidelines and historical data. Therefore, giving plant managers the best chances to keep machines in good repair. However, it still requires cyclical planned downtime.

However, preventive maintenance is estimated to save companies around 12–18 percent in costs compared to reactive maintenance.

What is predictive maintenance?

Predictive maintenance, on the other hand, only occurs when needed. It relies on real-time data from IIoT-connected equipment to identify potential threats before a problem occurs. In this way, repairs address an actual problem and are more targeted. This means that downtime, when required, is reduced by 25 - 30 percent compared to other maintenance methods.

Storing big data

To work effectively, predictive maintenance relies on data from sensors that report on equipment’s health state. However, IBM estimates that about 90 per cent of all data generated by sensors goes unused. This means that manufacturers miss opportunities to make informed decisions about their equipment while still paying to collect and store data.

Data collected but not processed or used in any way is known as dark data. It represents a sizable challenge for the industry.

Sensors to collect data can be relatively inexpensive and easy to set up. However, the challenging part is processing the data to draw conclusions about the health of the machine.

Processing data can be challenging for many reasons, from understanding the data to placing it with the relevant department. For instance, data silos happen when data is processed and relevant patterns discovered. However, the different departments of an organisation do not share these patterns.

This can happen if the business does not have the necessary technology for data visibility in place. For example, it might lack a unified integrated data management (IDM) tool or computerised maintenance management system (CMMS). This would mean every team might rely on different platforms.

Current transformers can collect raw data on an electric motor's eccentric rotors, winding issues, and rotor bar issues. However, without CMMS software, data may not be shared with the maintenance team on the production floor. This could result in unplanned downtime despite the use of sensors.

Lending a helping hand

Having sensors to gather information on equipment is the first step to incorporating the predictive technique into your maintenance strategy. Recent machines usually come with different options for real-time data acquisition.

However, legacy equipment can also be retrofitted with inexpensive add-on sensors. In fact, predictive maintenance can be a vital asset. It is particularly useful when dealing with ageing equipment. This is because ageing equipment requires careful planning when sourcing obsolete spare parts.

However, the industry contains an impressive amount of dark data. Coupled with the issue of data silos, this shows that gathering data is not enough. To predict equipment failure effectively, manufacturers should implement technologies that facilitate real-time data processing. Allowing all relevant personnel to access the resulting insights.

To combat these issues among different departments, businesses should prioritise the merging of information technology (IT) and operational technology (OT). The separate management of these teams from different skill sets was once logical. Yet, the increased digitalisation of manufacturing processes, including maintenance, means that it is now necessary to bring these areas together.

OT collects raw data from PLCs, sensors and other key equipment, while IT gives the data meaning by uncovering relevant patterns. However, both equipment and teams must communicate and collaborate proactively.

One of the issues with data collected by industrial equipment is that as it ages, its relevancy and accuracy decline. In this sense, edge computing can be a valuable solution. Edge computing minimises latency and supports real-time decision making by analysing data as close as possible to the source. This is rather than sending it all to the cloud.

Edge computing can also assist with connecting more equipment to the IIoT—cybersecurity. When data travels back and forth from the cloud, it increases the risk of compromising it. Processing data closer to the source reduces this risk.

It offers the advantages of increased digitalisation without opening up more potential attack surfaces. This does not mean you should avoid processing or storing data in the cloud at all costs. It simply means that the two options can go hand in hand to maximise results.

Predictive and preventative maintenance will always be more successful than reactive maintenance. Even, despite the problems with dark data and the necessity for a cultural transformation in manufacturing operations. The reduced downtime, costs and increased efficiency are what manufacturers need to thrive in an increasingly digital world.