Predictive Maintenance Strategies
How this Approach is Changing the Way Chemical Plants Function
- By offering early fault detection, better safety and improved plant efficiency, predictive maintenance is changing the way that chemical plants function. © Zora Zhuang/Getty Images
- The SAM4 predictive maintenance dashboard recommending replacement of a cardan shaft for a specific asset. © Semiotic Labs
- Jack Saville, Semiotic Labs, Leiden, The Netherlands
Unplanned downtime of critical equipment is bad news for any chemical plant owner. It can result in missed output quotas, higher maintenance costs and even threats to employee safety. The traditional method of reducing unplanned asset downtime is through preventive maintenance, where maintenance is scheduled after a certain period of time, regardless of whether a fault has been detected. The thinking behind this was that all assets will break eventually, so it’s better to maintain or even replace them after a certain period of time.
However, driven by a need to more effectively reduce unplanned downtime, improve plant efficiency and increase general plant safety, many plant owners are ditching preventive maintenance for a more data driven, and ultimately more effective strategy — known as predictive maintenance.
Reduce Unplanned Asset Downtime
Many chemical plants have realized that preventive maintenance is not necessarily an effective method of reducing unplanned downtime of critical assets. This is because the preventive replacement time is determined by when the asset is expected to break (based on experience), and not actual insight into the condition of the asset. Consequently, assets often break down before the preventive replacement of the asset takes place.
Predictive maintenance is more effective at reducing unplanned downtime. This is because it involves the collection and analysis of actual motor/asset data to help predict when the specific motor/asset will fail. The maintenance engineer will be warned of asset failures up to five months in advance, which gives the engineer sufficient time to schedule downtime and conduct maintenance. Some more advanced forms of predictive maintenance, such as SAM4 by Semiotic Labs, are able to detect over nine out of ten developing faults, therefore giving plant owners a much more effective way of reducing (and even eliminating) unplanned downtime.
Improve Plant Operational Efficiency
Predictive maintenance can also enable chemical plants to improve operational efficiency. There are several reasons for this. Firstly, and as mentioned above, predictive maintenance is a more effective way of reducing asset failure and the downtime that follows.
Crucially for plant operations, unplanned downtime can result in missed output quotas. By predictively maintaining assets rather than preventively, unplanned downtime can be better avoided — giving plant owners a better chance of hitting output quotas.
MCSA (motor current signature analysis) driven predictive maintenance can also help your chemical plant to reduce electricity consumption. MCSA based solutions measure the current and voltage of critical assets to determine the presence of a fault, whether a motor needs to be rightsized for its process, or whether the process is inefficient by design. By acting on this information, maintenance teams can reduce asset electricity consumption, therefore reducing plant energy consumption and ultimately reducing plant operating costs.
Predictive maintenance also enables increases in overall equipment effectiveness (OEE), an important measure of operational efficiency. If assets are only replaced or serviced when an actual fault has been detected, then healthy assets can continue to run until it’s necessary to change them. Within a preventive maintenance strategy, healthy assets are often unnecessarily replaced, which decreases OEE and increases operations costs.
Additionally, a predictive maintenance strategy also enables a more efficient use of maintenance resources. This is because maintenance personnel are only dispatched when an asset fault is detected, and not before. This lightens the maintenance workload, which subsequently reduces operating costs for the plant.
When a fault is detected, some MCSA based predictive maintenance solutions can even make the act of maintenance more efficient, and consequently decreases the scheduled plant downtime window. By using motor data, solutions such as SAM4 are able to determine the actual reason for the fault and offer recommended maintenance actions. This means that when the engineer inspects the motor, he or she knows exactly what to look for, and does not need to conduct a full asset inspection.
Improve Plant Safety
Predictive maintenance also strengthens plant safety. In the chemical industry, critical asset failure can put employees in harm’s way. For example, if a pump transporting a chemical breaks, then that may result in a hazardous chemical spillage. Predictive maintenance is the most effective way to reduce asset failure, which can consequently reduce the risk of safety incidents.
By conducting necessary maintenance only, instances in which a maintenance engineer needs to visit assets operating in hazardous locations in the plant is also reduced. In a chemical plant, assets can often be placed in hazardous areas, and whenever an engineer needs to conduct maintenance in that area, there is a potential threat to safety. Predictive maintenance means maintenance personnel are only deployed when there is a problem to fix, and so unnecessary visits to hazardous locations can be avoided, and overall employee safety can be increased.
To sum up, predictive maintenance helps chemical plant owners to reduce downtime of critical assets, improve plant operations (for example by improving efficiency and increasing plant uptime) and improve plant safety. All these factors are driving many chemical plant owners to seriously consider adopting predictive maintenance strategies.