February 10, 2021
By: Lyat Avidor Peleg
Predictive Monitoring for Greater Value Impact
When it comes to enhancing value in process plants, increasing production has a greater impact than cutting maintenance costs. Although these costs are the first thing that comes to mind when talking about predictive maintenance, the true benefit of the technology lies in helping customers maximize production.
Predictive maintenance has been a buzzword in the process manufacturing industry for some time. It’s understandable; you’d be hard-pressed to find an ops team that doesn’t want to cut maintenance costs, which is one of the most obvious benefits of predictive maintenance tools.
With the help of predictive monitoring, maintenance teams gain advanced warnings about potential part failures that often allow them to repair equipment before it needs to be replaced, reducing costs for replacement parts, and to schedule maintenance and repairs for times that will cause least disruption.
Webinar: Introducing Predictive Maintenance as a Service Register
The serious impact of unplanned downtime
But talking about the benefits of saving money on replacement parts only scratches the surface of the potential value of predictive monitoring. Sudden part failure frequently causes unplanned plant downtime, as production has to be halted until maintenance teams can fix the issue. If replacement parts aren’t immediately available, downtime can stretch on for several hours or days.
According to research, 82% of oil and gas plants experienced at least one incident of unplanned downtime over the previous three years, at an average cost of $250,000 per hour. The average plant had two incidents of unplanned downtime, and the total cost amounted to over $2 million per plant.
Additionally, equipment that isn’t operating at peak performance drags down both production quality and quantity. By producing the earliest possible alerts about emerging anomalies within the plant, predictive monitoring systems can guide ops teams to inefficiencies within the plant that are harming production rates.
It’s clear that the amount of loss due to unplanned downtime and reduced plant efficiency far outweighs the loss due to elevated maintenance costs. We’ve seen many of our customers discover that increasing production efficiency by small fixes has the effect of increasing production by a few hours or even a couple of days across the year, significantly boosting revenue.
This value underpins Gartner’s recent hype cycle report about data science and machine learning, which concluded that the benefit of predictive analytics is high. As stated in the section of the report entitled “Climbing the Slope: Predictive Analytics,” “The excitement surrounding predictive analytics continues to drive more interest and adoption at all maturity levels. Levels of project underperformance and ROI failure are low and this technology has quickly crossed Trough of Disillusionment as the rate of evolution and underlying value of predictive analytics drives the technology rapidly toward the Plateau of Productivity in the near future.”
2020 emphasized the true value of predictive monitoring
The arrival of COVID-19 highlighted the value of predictive monitoring. With most process manufacturing companies forced to run plants with a skeleton staff and/or remote monitoring teams, it became even more imperative to receive early alerts about potential issues before they become serious. No plant can afford unexpected downtime when full ops teams aren’t on site every day.
At the same time, the global progression of the pandemic brought supply chain and distribution issues that challenged plants to become more competitive and efficient, in order to avoid losing market share. Boosting production became even more important.
Our customers discovered the real value of correcting small issues before they snowball. By addressing and correcting small, non-repeating problems, and calculating the estimated monetary impact of what could have happened (best and worst case scenarios), they calculated that they saved an average of $14.5 million over two months when considering the overall risk of NOT having fixed the issues.
What is Predictive Monitoring?
Predictive monitoring is distinguished from predictive maintenance because it monitors the entire plant, not just the most expensive or failure-prone pieces of equipment. Predictive monitoring solutions like SAM GUARD produce the earliest possible alerts about plant anomalies, which direct plant engineers to areas of inefficiency within the plant long before they cause part failure.
By correcting small inefficiencies, plant engineers can prevent shutdowns and ensure that the entire plant runs more smoothly, raising production quality and production rates.
In contrast, predictive maintenance tracks the performance of a small number of expensive and large items within the plant, helping ops teams estimate when they are likely to fail so that they can plan replacement and repair times. This helps reduce the expense of replacing a part too early, but does little to boost production or extend part life cycle.
At Precognize, we offer four main applications within our predictive monitoring capabilities that help our customers ensure smoother operations and inform them immediately of any problems. These include:
- Early detection of equipment failures. Prevent shutdowns before they occur by detecting future failures at an earlier point.
- Suspicious operations behavior. Alert the operations team immediately if the system detects any variations to the usual process.
- Suspicious operations modes. Alerts about any unusual operation modes due to employees who deviate from the planned process instructions.
Predictive monitoring can deliver on all these areas because it monitors the entire plant as a whole for any anomaly, rather than only tracking selected items of equipment.
Getting the Most Out of Predictive Monitoring
If you want to maximize the value of predictive monitoring, you need a clear understanding of your goals from the very beginning. We recommend establishing Key Performance Indicators (KPIs) to clarify your business goals and use cases.
- What do you want to improve within your plant?
- What is your measurable goal, in hard numbers?
- What is your plant’s current status? Establish a baseline for measuring future changes.
- Which equipment relates to this goal?
Predictive monitoring enables you to respond in real time to any issues that arise in your plant. It could be something as minor as cleaning a valve to allow smoother passage of raw materials, or as major as replacing a mechanical part. Rather than waiting for something to go visibly wrong, rely on a system that helps make production more smooth and efficient to raise plant value significantly.
Since predictive monitoring is still being adopted and needs to be woven into the workflow, it has a lot of potential to grow and improve. As more and more data is captured within plant historians, and an increasing number of plants apply this technology, predictive monitoring will advance to unlock ever more value within process plants.
Learn More about Predictive Monitoring
You can learn more about how to bring true value to your plant through predictive monitoring, by watching this webinar presented by the CEO of Precognize, Chen Linchevski.