Relationship Examples
Discover What You Can Do With Relationships
It isn’t always obvious just how powerful relationships really are.
Below are a few examples that highlight how we use them, but they’re also something that can be used creatively by anyone in your organisation.
Some examples are going to be unique and powerful. Some are going to be quite simple. But one thing to remember is that when ARDI is being used, all of these update with changes.
The main issue with any of these examples is…
Using relationships, neither of these apply. Anyone – even people completely unfamiliar with the system – can build a display or create an analytic without needing engineering support. And when you add a new assets (ie. vehicles, pumps, people etc.), they instantly appear..
Retiring a truck? No need to fix up the live display – it will automatically refresh with the right number of vehicles.
Allow people – including those who aren’t technical – to understand how the parts of a process are connected and work together, without having to open P&ID diagrams.
Allows people to navigate the system based on connection rather than just ISA-95 (which tends to dead-end). In a well-built ARDI database, you can follow assets from any point to either an input or output of the system.
Helps organise assets into human-readable, understandable systems and sub-systems. It also groups assets into types, which makes it simple to find similar equipment and maintain large systems.
Relationships can be used for smart deployment of solutions such as alarms, calculations and analytics. Instead of manually adding individual calculations for every valve you have, you can build the calculation once and deploy it to every valve in your system. It then keeps up to date as new valves come online or old ones are retired.
Helps staff determine when upstream faults/alarms are contributing to issues down-stream.
Helps automated tools/LLM models suggest possible causes and resolutions when trying to diagnose issues.
Helps identify what assets will be effected by switching off/isolating/modifying specific pieces of equipment.
Helps identify isolation points for a particular piece of equipment.
Allows the system to suggest related assets when interacting with the system. When creating charts such as scatter-plots, we can suggest all of the assets that will most likely be relevant (ie. If you select a pump, you’ll be able to very quickly compare it’s speed against other pumps in the same bank, vs current and voltage, pressure, overall system speed and conditions and environmental factors). This radically reduces both the time it takes and the detailed process knowledge required to perform analytics
Allows the runtime creation of zero-maintenance displays, dashboards and reports using only structural data.
Logical relationships (such as sequence, startup order etc.) can be used to detect unexpected spikes or troughs along the direction of the relationship. Can be used to detect leaks and blockages as well as out-of-balance load-sharing.
Provides data structure and context to LLMs so you can use natural language to query data, build visualisations and trouble shoot.
Gathers all variables that contribute to the overall context of not only what an asset is doing at the current time, but why. For example, if you look at a conveyor motor, you not only see the status of the motor, but the weight of the product on the conveyor, the type of product it is and the ambient environmental conditions – all factors that contribute to your understanding of the values being seen on the motor.
Allows you to perform flow-analytics using the down-stream path of water, power, gas, network traffic and other forms of consumption to identify consumers, blind-spots, inefficiencies etc.
When trying to perform analytics such as energy efficiency (ie. how efficiently the process is turning natural gas into heat), relationships can be used to determine which variables are associated which assets based on tracking the flow-path. For example if you have five ovens, five control valves and only one flow meter, you can determine individual oven efficiencies automatically by associating the valves with the ovens via their natural gas relationship.
When performing correlation analysis (determining what parts of the process are impacted by others), telling the difference between correlation and causation is difficult. Relationships give us some insight into if two systems are physically or logically connected, which helps us determine if causation is likely.
Allows you to present your data in a way that is natural to the operators and field technicians – for example, in process order rather than alphabetical order.
Automatic filtering of assets to specific areas, buildings, systems, sections, domains etc. You can build visualisations and reports for just the machines in Building A, just the components controlled by a particular PLC, just those down from a specific valve etc.
Helps direct people to the most efficient route between points – both physically (in terms of maps and 360 photography) and logically, in terms of distance in pipework or network diagrams.
Can be used in several forms of analysis to help appropriately allocate responsibility for issues. For example, in utilisation analysis to ensure downtime or issues are associated with the correct parts of the system. If a breakdown occurs at point ‘B’ in a process line with assets in the A -> B -> C -> D order, you don’t want to report assets ‘C’ and ‘D’ as being at fault for a process outage, even though they will be stopped.
The best way to understand the power of ARDI is to see it with your own eyes.
Simply add your contact details below and we will arrange a personal demo where we will take you on a full tour of ARDI’s backend and show live demonstrations of each capability