AF Accelerator FAQ

Our customers ask many questions about AF Accelerator and how we can help build and maintain PI Asset Framework (AF) models.

If you don't see your question below, please contact us at or use the 'Request Demo' link above.

Yellow Line Thin

AF Accelerator builds enterprise-ready PI AFs in as little as 30 days. The solution combines Element’s Forward Deployed Engineering (FDE) Team, Industrial Data Method, and AssetHubTM technology.

AF Accelerator overcomes the data integration, modeling, and maintenance barriers inherent in using spreadsheets to build and maintain PI AFs at scale. Our PI AF experts deploy Element AssetHub and use a proven industrial data methodology that accelerates the building of PI AFs 10x, increases the efficiency of your engineers by 4x, and helps enable critical operational insights faster and at a lower cost and risk than any other solution. Visit here to learn more.


We already have an existing PI AF built, but it's not complete. Do I have to start over? No, you do not have start over. We can ingest your existing PI AFs. We see this use case a lot -- customers want to either augment their existing PI AF, or take an existing PI AF and reshape it for a different use case. This allows them to reduce redundant work, and move faster in their development and deployment.
What if my tags don't contain context? That's OK. There are a few different design paradigms that we implement: First, when there is context in the PI Tags; Second, we might need to pull in the Enterprise Asset Management (EAM) system metadata; Third, bringing in a spreadsheet containing P&ID context, and lastly; encoding Asset Engineer's knowledge of their tags into spreadsheets and uploading this to AssetHub.. Once you have an Enterprise Asset Model built in AssetHub, you also get additional benefits including an unified data model store and a way to quickly deploy PI AF hierarchies for future use cases.
Once we build PI AFs, if we need to unplug AF Accelerator, can that be done as we like to be self-contained ? Yes, absolutely you can. You can walk away with the PI AFs that you built and not have to use Element again. However, we're confident that AF Accelerator has additional features to help you manage and maintain your models and PI AFs long term.
If all we have are just spreadsheets, do we still enjoy all the features you have, or is your tool hampered in some way? No, AssetHub is not hampered. It works well with spreadsheets.
Can we use our PI AF template files ? Yes, we can import your current PI AF templates as spreadsheets and use them to build your asset data models.
Do you have a template library that we can use to say, model our cyclone feed pumps or other assets?We have almost 100 asset templates defined in our library. If your equipment class isn't included, we can work with you to build a new template. Or you can take a template that is a close-fit to your asset, tweak it, then save that to library so others can use it at other sites, or for different applications.
How do we add functional location work orders from our ERP (SAP and SAP PM) system so we can have a more complete view of our asset? And are updates synchronized automatically back to the model? We use software agents and other methods to connect to those systems to pull in the contextual data to allow AssetHub to integrate that within the Enterprise Asset Model. If you are an OSIsoft PI customer, then we can export that model as a PI AF. If something in the source data changes, because we maintain a connection to the source, the model is continuously kept up-to-date. If someone needs to add, say, 'cost center' to the PI AF, you can integrate that into the model, then create a new PI AF.
How do you extract or manipulate data taken from another system/non-time series?Once its been ingested via the AssetHub data input portal, you drag it onto the pipeline canvas where you begin transforming the data. The pipelining process uses map, join, and parse (and other) functions to manipulate the data and contextualize the time-series data with data from those other systems - building your Enterprise Asset Model. You then choose the hierarchy that fits your needs, and export that hierarchy as a PI AF.


When you extract time-series data, does data leave the PI server? Time-series data stays on the server - we only load the PI Tag Configuration into AssetHub. We never change the source system. The data copied is used for data transformations upon export and for ongoing data quality checks - such as running integrity reports to detect tag coverage gaps for an asset.


How do you deploy your software? Element AssetHub is deployed on Microsoft Azure. Our customers typically prefer to host AssetHub in their own Azure tenant, but if they do not have Azure, we can provide the hosting.


What is your relationship with OSIsoft? As an OSIsoft Connected Application partner, we are committed to OSIsoft's customer base, ensuring industrial organizations garner maximum value out of their PI System by assisting users in building robust PI AFs 10x faster than using traditional approaches using spreadsheets. Element was founded in 2015 by Sameer Kalwani and David Mount. Sameer split his time between customers and OSIsoft's HQ, narrowing-in on the opportunity. Ex-OSIsoft employees also work at Element - providing decades of PI System experience working with customers time-series data and PI AFs.
How is AssetHub related to AF Accelerator? AssetHub is the core technology of our AF Accelerator solution. AssetHub connects to disparate asset data sources, manages data modeling to build Enterprise Asset Models (the dynamic digital representations of physical assets), and shares that common model, and asset data, for easy consumption across the enterprise as a PI AF. AssetHub is delivered on a modern, cloud-native architecture to overcome the integration, data modeling, and persistence challenges inherent in utilizing asset data for advanced industrial analytics.


Are there any scaling limitations ? We have customers using AssetHub in production that have 1M+ tags in their PI System. When you consider all the metadata associated with each PI tag, the number of data connections increases exponentially. AssetHub is built on top of a very high performance compute engine, allowing users to build data models without worrying about the speed of existing data entry tools.


Tell me about security AssetHub is typically deployed on your Azure tenant so your data does not leave your private network. We can also host AssetHub on our tenant in a dedicated instance. When building PI AFs, we only use the PI tag metadata, and not the actual time series data, ensuring that the process data stays within your network. We use encryption at rest and in motion so you can trust that your data is secure.

Next steps

Is there an AssetHub demo I can view before I contact you? Yes! We shared a demo in our latest webinar here