Questions & Answers

Hello, is there a way to navigate to multiple dashboards from a single source, without the use of MLHub?  Example: I have a primary dashboard and on it, I can click a tile and move to a different dashboard similar to how we can transition to TrendHub directly via the tile or trend view.

Hi,We have a continuous process with around 50 product types.For all of those product types I have different KPI’s e.g. for steam consumption / steam flow.Which trendminer feature is able to compare current consumption at product type xy vs. standard consumption at the same product type? I would like to have a Dashboard showing standard consumption vs. current consumption and a green, yellow, red flag (depending on %-value of lower / higher consumption vs. standard).I have a list of product types and corresponding standard consumptions, but I do not know how to use as a comparison reference.Thanks for your help

Hello,I imported Tags into trendminer (first trial, I am a trendminer beginner).I recognized, that the timestamps are not good and deleted all imported Tags.Now I want to import the same Tags again (with optimized time data), but get the following message:According to this message, the Tags already exist.When clicking on “importierte Tags”I get this:Its contradictory to the message “Tags sind bereits vorhanden”What can I do to really delete the Tags and to import them again (with different time data)?Thank you for your support

I was able to follow the instructions on another thread on how to create an aggregation tag for an integration over 24 hours of a binary tag for line running or not and it will give me the downtime over the last 24 hours at any specific point. Is there a way to lock in a time so that on any given day at 7 am, it will show me the downtime over the last 24 hours to the previous day at 7 am? We have a few metrics like that that we report out on over the 7 am - 7 am window and that feature would be very helpful.

Hi allIs there a way to select different key parameter for every row, when I do a layer comparison?I would need “Delta” for the first row and “Average” for the second rowBR Peter

When a monitor is created to detect deviations of a given fingerprint, in the monitor gives an score on how adhered was the production run to the fingerprint, the difference is the deviation, how could this score be registered in a context table? Score of fingerprint monitorThanks.

Discrete data are automatically interpolated, is there a solution to only see the real data and export it? 

Hi,I’d like to load the same tag multiple times in the same view. I then want to give each of these duplicates a separate time shift so I can visually compare different periods. Is this possible?Thanks.

Hello,we are working on a use case where we would like do do a re-sync of historic items for a longer period. To go via the interface for such a long time or performing a full historic sync don’t seem to be the best options. We could imagine a few possible ways to accomplish this task:Python / powershell script Swagger  API calls such as ‘12.22.3. Re-sync list of context items’ and run these via tools like postmanCould you help us with setting up a way so that we can do the sync ourselves when needed? Here are a few infos regarding our case:SaaS instance Postman, VS code available Context Item Type: BATCH ANALYSIS Time frame: 1.1.2022 - today Expected number of Context items: 5000Thank you for your help already!Best​​​​​​​Sören 

Hey,we observed that fingerprints created on tags can not be used for their correspondet attributes. For example:The pressure of a reactor vessel is called A100P01/MESS.XABS and mapped to the attribute ‘Reactor pressure’. When a fingerprint for the reactor rpessure is created using the tag name, the fingerprint can not be displayed when the Reactor pressure is added to the view as the attribute ‘Reactor pressure’ although the attribute references the tag name in the attribute. I’d consider this a bug since the link between attributes and tags is given in the AF structure.  BestSören 

We're looking to optimize our TrendMiner implementation and need expert guidance on advanced industrial analytics best practices. Seeking a consultant with 1-2 weeks availability to help us:Develop comprehensive time-series data analysis workflows for process optimization and root cause analysis Set up advanced pattern recognition, predictive analytics, and asset performance monitoring frameworks Integrate TrendMiner with our existing process historians (PI, IP.21, PHD) and contextual data sources (LIMS, MMS, OEE) Optimize custom calculations, event analytics, and production cockpit dashboards for operational decision-making Train our process engineers and operators on self-service analytics and continuous improvement methodologiesWe're in process manufacturing looking to leverage our operational data for better performance insights and proactive maintenance strategies.Experience with TrendMiner implementations, process manufacturing analytics, and Industry 4.0 initiatives preferred.

Hello, I would like to make a monthly average over a TAG. Since the maximum aggregation period is 1 day I would like to know how to make this calculation using another method. I already checked the topic How Can I Apply Stronger Aggregation for Long-Term? | TrendMiner Community but this does not answer as I need to save this calculation as a new TAG. Please let me know how to solve this. Thank you,Victor Nassau

I need help to set up a monitor and an automated e-mail when a tag goes bad.Not when the the tag goes to zero or any other defined value, but when it’s missing completely.

Hello,Is the score of a deviation in any way affected by the scale settings of a tag when saving the fingerprint? For example: We are saving fingerprints and are changing the range of the y-axis for visualisation reasons. Therefore, we would like to know whether these changes have an affect on the threshold for fingerprint deviation monitors. The documentation states:“Scores for similarity search results are based on visual similarities between the query (solid line) and the search result (dashed line). At every point the absolute difference is taken, summed, and rescaled to the range of the query time period. A relative score from 0% to 100% is obtained.” Hence, I am wondering whether the scale has an impact on the fingerprint deviation scores too. BestSören

Hello, I’m looking for a way to setup of formula to return value of tag A during certain condition based on value of another tag B. The goal is to check, if during specific phase (“phase” is when tag B is in the range of 1 and 10), the value of A tag exceeds the value captured during the start of the phase, for, let’s say, 5 points increase.I assume this all can be done setting up using formulas and monitoring, but the bottleneck currently is to capture the start value of the tag during the start of the phase and to reference to this value during the phase duration (and this logic should be happening with each new phase, indeed).Example for better understanding: tag B becomes 1, we need to capture the value of tag A at this moment (let’s say 3), and then during all the time when tag B is going the way from 1 to 10, we need to constantly check if the value of tag A is not increasing on more than 5 points (meaning, threshold for this specific phase would be 3 + 5 = 8); so if value of tag

I am doing a quality analysis for which I have created specific trends. Now whenever there is a new quality issue, I can use these trends to analyse the problem. However, since the issues occur on different times and dates I need to change the basis layer and the reference layer for correct comparison.  I created a dashboard with all the relevant information. The information is very simple. It shows multiple graphs of KPIs with a basis layer and a further reference layer for comparison. To save me some time, I was hoping that it is somehow possible to change the date and time of both layers for all of the trends at the same time. Is this possible and if yes, how?  

Good day, TrendMiner hive.Our IT department has approached me about optimizing our on-prem Azure storage allocation. It appears that our current setup may be over-provisioned for our actual usage.According to the online documentation for 2025.R2, the minimum storage requirements are listed as:SSD, 200GB + 300GB, 250MB/s.However, our current usage is significantly lower:OS: 21GB used out of 511GB provisioned Storage: 109GB used out of 600GB provisionedGiven this, IT is considering reducing the allocation to something more aligned with our needs—perhaps 64GB for OS and 200GB for Storage.Would such a configuration be sufficient for running and upgrading without issues?Any insights or recommendations about adjusted the storage footprint would be greatly appreciated.Thanks in advance!— Edward Burch

I created a calculated load of an adsorption bed via custom calculations.Unfortunately the value of the load resets every beginning of a new month. I guess that’s related to the index granularity.Is there any chance of continuing the values from the previous month?Best regardsSascha

Is there a way where you can calculate the total production of a certain product. What I mean is that you consume this certain product numerous times a year. But every time the totaliser gets reset and starts again at zero. I want a way where you can combine all the totalisers so you know how much you consumed during the whole year.Example:Product A: 0 to 10000Product A: 0 to 150000Product A: 0 to 20000End total 2025 Product A: 180000 kg 

Dear Team,Can we save the view with future dates let’s say 4 days future data and use this view in dashboard? Currently we are making live view which refresh and show up to current timestamp. what if we want to use this as current+4days?

Hi,  I am trying to set up a Custom Calculation to check if a variable is under many conditions, but I got stuck in a problem. The two limits (max and min) only get new data points in the data base when we change recipes. I tried to use .ffill() to fill the data base with the older values, but it did not work. I still get NaN results for the time I dont have values.   

Dear Team,We have value based search for tag. We want to do value base search and trigger alarm for future value as well. How it is possible?

Dear Team,How to set the frequency of email in monitoring tool?

I just found these tags in TM: TM_hour_UTC, etc. Is there any documentation on these? This is a builtin set of tags, or a specific datasource? And why do we have hour, but not minute? Because of the 1m index resolution this might not make sense?