Use Cases

Most plants run more monitors than anyone can review in a week. Across three sites, that can easily mean 400 monitors and well over a thousand triggers.With our TrendMiner MCP integration, an AI assistant can already read your monitors and their trigger history directly. That makes work that is close to impossible by hand a routine task:A weekly report of every monitor that fired, across all your sites A ranking of what to look at first, based on how often a monitor fired, its trend and how critical the asset is A list of noisy monitors that fire constantly and probably need retuningThe AI agent does the scanning, and your engineers decide what to act on. 

Condition-based maintenance 🔧 is powerful. Monitor what's happening and act only when needed. But some maintenance tasks still need to track time 🕑. Regulatory requirements don't care if your asset is running smoothly. Some failure modes are unpredictable and preventive by nature. And if your equipment runs 2 hours in winter and 18 in summer, a calendar-based schedule is useless. You need to track actual operating hours since the last maintenance event.The strongest programs use both. You monitor bearing temperature and oil quality in real time. You also ensure critical seals get replaced every 5,000 operating hours. With TrendMiner's custom calculation feature, you can build a time-since-maintenance tag by defining the operating conditions that "count" toward your interval, then tallying the actual hours. Set a monitor and get alerted when you hit the threshold. When the equipment has actually earned the maintenance.Want to build operating-hour counters for your equipment? Reach out

 BackgroundEvery production site collects and stores massive amounts of process data — far more than is typically analyzed. For engineers, gaining a holistic view of plant operations can be a daunting task.TrendMiner’s layer comparison and compare table features make it easy to visualize and quantify how different operating periods performed, side by side.By leveraging this functionality, users can quickly identify high-performing periods and understand which process conditions led to those results. ObjectiveUse historical data to identify ideal operating conditions by comparing performance over several time periods (for example, across months).With just a few clicks, TrendMiner users can: Compare key process variables over multiple months, Identify which periods performed best, and Spot potential process differences that may have led to underperformance.  Solution1. Create a Layer for Each PeriodStart by selecting the process tags you want to compare — for example: Raw material fl

📣 If you missed our March 24th webinar, Optimizing Batch Processing Time and Increasing Daily Production with TrendMiner, the recording is now available and well worth your time. The session featured a compelling real-world from Mallard Creek Polymers, a specialty latex manufacturer, showing how their team used TrendMiner to reduce stripper cycle times by 10-20% and directly increase daily production output.Even better, an analysis that could have taken an engineer hours or days to complete, now takes about ten minutes to repeat.Find the webinar recording here 🔊  

One of my favorite uses for the multi-layer scatter plots is to check performance before and after asset maintenance. Not only is it interesting to compare how much the situation improved by maintenance, you can also see at a glance whether your maintenance is always equally successful, and whether you are performing it at a consistent level of degraded performance.This scatter plot shows the correlation between product throughput (x-axis) and specific energy consumption (y-axis) for 1 day periods just before and after 3 maintenance events. I picked the same color for the layer just before and after each maintenance event, so I can see what layers belong together. What we can indeed see from this plot is that the specific energy consumption decreases after maintenance, always returning to approximately the same level. This indicates maintenance was done successfully each time. We should also notice though that we clearly do not perform maintenance at the same level of performance degra

 BackgroundIn many chemical plants, reactors contain catalyst beds that gradually lose activity and need to be replaced on a routine basis. Because catalyst life can vary significantly, it is difficult to predict exactly when a failure will occur.A premature catalyst failure would have a major impact on production, supply chain, and maintenance. Therefore, it is critical to plan preventative maintenance well in advance – at least a few weeks prior to failure – to avoid costly unplanned downtime and production losses. ChallengeWith TrendMiner, a workflow can be set up to monitor and predict catalyst performance: Searching on Process Values The first step is to use value-based searches to identify when process values reach certain limits. These searches can be saved and later used as the basis for monitors. In this way, early indicators, warning levels, and critical conditions can be defined, all providing valuable insights into catalyst performance. Event analytics can of course also be

BackgroundInterpreting time-series data at a glance can be challenging, especially when process experts need a quick update on what’s happening in their plant. Instead of scanning raw trends directly, it’s often more effective to use a dashboard that highlights key incidents and events. With TrendMiner, dashboards can be configured to automatically track setpoints, rapid process changes, error codes, and much more, providing immediate insights without the need for manual data review. As an example, we will set up a productivity tracking system to monitor sulfur production in a sulfur recovery plant.Sulfur recovery plant dashboard. Workflow1. Set up monitoring on interesting sensor data2. Create a new ContextHub view with corresponding filters3. Add a Gantt-chart tile on a dashboard.4. Configure new Context Types (Optional; Admin rights)--------------------------------------------------------------------------------------------------------------------------------------------1. Set up mo

Produced batches often need to meet specific quality criteria depending on the product grade. In TrendMiner, dashboards can be used to monitor the different types of batches produced. For example, batches can be categorized by parameters such as product grade or quality. This provides a clear, real-time overview of how many batches of each category were produced over a user-defined time period. As an example, we’ll use lab quality data from a polymer production process. The plant produces three different product grades, and quality is evaluated based on viscosity measurements. WorkflowCreate a new ContextHub view Add a ContextHub counter tile Repeat for multiple trackers 1. Create a new ContextHub viewThe content of a ContextHub list can be customized using a variety of filters. Common filter options include time period, fields, tags or assets, types, and keywords. For our specific case, we are going to select:Period filter: Event open in range from 01/01/2018 until 01/01/2020 Field fi

 BackgroundIn batch processes in the process industry, quality fluctuations occasionally occur—for example, in the form of low concentration in the end product. These "bad batches" lead to increased waste, uncertainty in process control, and additional analysis effort. Often, it is unclear what exactly causes these deviations. ObjectiveThis article demonstrates how TrendMiner’s Event Analytics can be used to identify the root causes of poor batch quality. Value-based search, calculated KPIs, and graphical evaluations are employed. ChallengeHow can all relevant batches of a product type be systematically identified, meaningful KPIs calculated, and differences between good and poor-quality batches visualized using Event Analytics? Solution1. Load tagsFirst, load all relevant process data (e.g., pressure, temperature, concentration, fill level, dosing behavior) into TrendHub. Example of a TrendHub view 2. Define time intervalLimit the time window to the relevant period to focus the analys

Improve asset availability and energy efficiency with self-service analytics BackgroundIn many process-driven industries, heat exchangers play a crucial role in maintaining efficient thermal operations. For example, process gases exiting a shell-and-tube heat exchanger are often routed to downstream equipment like baghouse filters, which are subject to strict temperature limits. To prevent safety risks or equipment damage, conservative temperature thresholds are often used—typically set well below actual specification limits.However, over time, systems may experience efficiency losses due to fouling, which can increase outlet temperatures and push operations closer to critical thresholds. Being able to detect and address fouling at an early stage is key to maintaining asset availability and optimizing energy use. ChallengeFouling inside heat exchangers reduces thermal efficiency, leading to higher outlet temperatures and potential process deviations.To ensure safe and optimal operation

Knowing how certain settings impact our main KPIs is key in fine-tuning a process to driveproductivity. Such a feat can only be accomplished by actually measuring the relevant KPIs in acomparative study.In this use case, you will see how you can very quickly assess the impact of process conditions on agiven KPI during a selected period. process conditions: equipment on/off, product grade, day of the week... KPI: energy usage, setpoint deviations, total production, ... period of choice: the period you currently have on your focus chartAn example would be assessing the impact our Advanced Process Controller (APC) has on the averagedeviation of our Process Value (PV) from our Setpoint (SP). Is it actually doing its job in reducing theaverage deviation? Step 1 - Defining the KPIAssuming we don't already have a KPI tag, we will need to use a formula to calculate the KPI we want.In the case of APC performance, a typical KPI would be the absolute value of the difference between a process valu

TrendMiner can be used to automatically track the status of our sensor data during a batch process. To prevent losing time continuously having to check process trends, we can make a dashboard telling the user immediately the status of the production process during a period of choice. This can also detect problems, which were missed before and have had a big impact on, for example, product quality. As an example, we will look at the production of an active pharmaceutical ingredient in a batch process. To show the principle, we are going to monitor if a batch is currently running, and subsequently we are going to check whether the temperature does not increase too much during the batch. WorkflowSet up monitoring on timeseries we want to track Create a new ContextHub view with corresponding filters Add a Gantt-chart tile on a dashboard Optional for admins: configure the context item Set up monitoring on timeseries we want to trackFirst step is to track the events we want to have on our da

On a daily basis, a process expert evaluates key process parameters. In TrendMiner, a process expert can easily populate a dashboard with the preferred parameters making your most important process parameters available with one click of a button. As an example, we are going to visualize the general status of a heat exchanger on a dashboard for the last 24 hours to immediately follow up on important process changes. WorkflowSaving important trends in the work organizer Creating a TrendHub view tile on a dashboard in DashHub Creating a traffic light tile on a dashboard Saving important trends in the work organizerFirstly, we need to save the correct views we want to put on our dashboard. This can be done in TrendHub where we can save the view that is visualized.For our case we are going to:Load the tags related to the heat exchanger: TM-T200.* Visualize the outlet temperature of the product, and the inlet and outlet temperature of the heating medium. Put the focus chart on 24 hours, fix

 Extending the concept of hourly averages to shorter time intervalsRelated post: Calculations of hourly average consumption values BackgroundIn process industries, consumption monitoring of resources such as steam, water, electricity, and compressed air is key to ensuring energy efficiency and sustainable operations. While hourly averages provide a solid overview, shorter intervals like 30 minutes can deliver even more granular insights—especially for fast-changing processes or detailed performance analysis.👉 In some cases, half-hourly averages are required for regulatory reporting, for example to comply with environmental or energy-related limits imposed by authorities. Having these values available ensures transparency and compliance. ChallengeHow can we calculate average consumption values for fixed 30-minute intervals and distinguish them from rolling values? And how can we ensure the values are visualized cleanly as stepwise changes in the trend? SolutionRolling 30-minute average

 BackgroundIn any industrial process, there are typically periods of optimal (good) and sub-optimal (bad) performance. Identifying and comparing these periods helps in understanding the differences and optimizing operations. By analyzing the variations between good and bad periods, process engineers can uncover key factors affecting efficiency and quality. ChallengeOne of the main challenges in process optimization is recognizing the factors that differentiate a good period from a bad one. Some common difficulties include: Identifying relevant time periods or batches in continuous and batch processes. Establishing objective quality metrics to categorize periods. Effectively comparing periods visually and numerically to extract actionable insights.  SolutionTrendMiner provides a structured approach to compare good and bad process periods through the following steps:1. Feature Search Use the value-based search function to find operating periods or batches. For batch processes, an o

For rotating equipment, we need to seal off moving components with mechanical seals that use a barrier fluid. There is no way to seal off moving components perfectly so there will always be some leakage. If product leakage to the environment is unacceptable (typically for SHE reasons), a zero emission seal needs to be used. For zero emission seals, there can only be leakage from the barrier fluid into the product, which is why a barrier fluid that is under higher pressure than the pumped fluid needs to be used that is compatible with the product. After all, at least some of this barrier fluid will end up in the product. A classical example of this is the usage of N2 as a barrier fluid to natural gas.Mechanical seals are common but critical components, and they can be monitored quite easily in TrendMiner as the monitoring criteria are simply value-based. We can sometimes use a value-based search directly, but if we need to compare two measurement (e.g. to ensure a difference in pressure