Home / Features / Perfectly predictable

Perfectly predictable

Peter Smith, Head of Business Development at IAconnects, looks at how a sensor-based predictive maintenance set-up can help prevent costly equipment failure before it happens

Conventional wisdom suggests that maintenance should be carried out in a uniform manner and at regular intervals. However, this fixed method of working doesn’t take into account how and when business assets are actually being used. What if you could monitor equipment and accurately predict failures before they occur?

This may sound like wishful thinking, but the latest sensor-based technology, paired with a building management system and linked via a gateway, can make this futuristic concept a reality in the here and now.

Many facilities or plant managers will be able to recall occasions when equipment has reached end-of-life status earlier than expected, often resulting in costly downtime and a long, drawn-out wait for replacement parts. Even if rigorous ‘preventative maintenance’ schedules are put in place and followed religiously, with all replacements based on the manufacturer’s product lifetime guidelines, unexpected failures can still occur – and seemingly at random.

According to research, only around 18 per cent of assets actually fail due to anticipated age-related issues, while 82 per cent of asset failures occur randomly. What this really means is that preventative maintenance isn’t particularly effective when it comes to preventing breakdowns. If preventative maintenance isn’t proving to be cost effective, some may decide to simply do away with strict maintenance plans altogether and adopt the ‘fix it when it breaks’ approach. The problem here is that when that equipment does fail, the fix usually takes a lot longer and typically ends up costing more.

The ideal scenario, therefore, would be one in which your equipment tells you when it needs to be maintained. Predictive maintenance, sometimes referred to as condition-based maintenance (CBM), makes use of monitoring technology in order to ensure that maintenance decisions are based on a genuine need rather than more arbitrary criteria such as the age of the asset.

Integrated sensors can ‘listen’ for potential problems and bring about a timely intervention before a complete failure occurs. The latest state-of-the-art sensors, carried in on a wave of Industry 4.0 innovation, can monitor for changes in temperature, vibration levels, output and more, sending the data to the cloud where IoT solutions use scalable monitoring, visualisation and predictive analytics to gain valuable insights into operations. As the data from sensors can’t be passed directly to the cloud, a gateway is needed to secure transmission and provide the necessary connectivity to other devices and protocols.

As well as obvious uses in manufacturing environments or in any building containing industrial plant equipment, other building assets with links to facilities can be fitted with sensors in order to maximise data collection and optimise maintenance procedures. Thousands of sensors can be deployed within a single building and as they can be retrofitted to almost any device or system, vast amounts of data can be collected in a relatively short space of time. This data then helps to clearly identify the assets that are in constant use or suffering the most wear and tear, thereby heightening the risk of short-term failure.

Once the connecting gateways have been set up to deliver the sensor data to the cloud, actions and alarms can be set up to trigger if and when certain conditions are met. This rule-based approach requires an understanding of the contributing factors to asset failure, but once the elements that require measurement have been agreed upon and the predefined levels set, the system is ready to deliver alerts to the operator as soon as the specified threshold is passed.

Correctly implemented, a sensor-based predictive maintenance arrangement should let the facilities manager or building operator know if there are any behavioural anomalies or if the equipment is likely to fail, as well as offering insight into its expected remaining life. By examining usage history, the operator can also gain insight into usage patterns and make adjustments in an attempt get the most from each asset and maximise efficiency.

Making full use of IoT-enabled technology and advances in smart building controls means finding new ways of capturing and using data. Predictive maintenance is part of a wider trend towards gaining full control of assets and maximising their potential, ultimately reducing operating costs and downtime.

By using a gateway which allows connectivity and control of wired and wireless devices via a 3 or 4G managed mobile data service, data from sensors can be captured, processed and used to transform outdated and ineffective maintenance procedures, giving users previously unimagined powers of prediction.

About Sarah OBeirne

Leave a Reply

Your email address will not be published. Required fields are marked *