Affectli brings every asset, site, person, event and piece of work into one live model of your operation. Any data point can be pivoted with a click, so every decision starts from the full context.
An operational ontology is one live model of everything your organisation runs and does. Each real thing is an object, with properties (what is known about it), links (how it relates to other things) and actions (what people can do with it).
Treatment works · aerialEvery basin is an object
Most operations already hold this knowledge. It sits in control-system (SCADA) tags, GIS maps, asset registers, ticket queues, spreadsheets and the memory of experienced staff, each with its own name for the same thing.
Affectli joins it into one live model that updates from the systems you already run. Every alarm then arrives with its asset, its site and its crew, and a question about one object is answered with the context of all of them.
| System | How it knows the reservoir |
|---|---|
| SCADA / PLC | RES14_LVL_PV |
| GIS | WTR-RSV-000014 |
| Asset register | A-0040112 |
| Ticketing | "Hilltop Res." (free text) |
| Spreadsheet | Row 214 · Hilltop SR |
Which homes lose supply if PS-07 stays down for four more hours?
Which reservoirs will reach critical level before the morning peak?
Which hospitals lose power, water or mobile coverage if substation SUB-0417 trips?
Take one reservoir. Its type says what kind of thing it is. Events keep it current, and permissions decide who can see and act on it. Hover over a part to find it on the object.
Reservoir, Pump Station, Engineer, Field Ticket. You define the types your operation needs, and every object of a type shares one structure.
Level, capacity, inflow, pressure, install date. Values arrive from the systems that hold them, many of them live.
A pump station fills a reservoir. A ticket is assigned to an engineer. Follow any link to pivot from one data point to the next.
Raise a field ticket, send a summary, start a workflow built in Affectli Designer. Actions run from the object, so the work starts with its full context.
Time-stamped occurrences from sensors, systems and people. The Mi-Stream engine applies rules, filters and integrity checks, then pushes events to the Events Monitor.
Set by classification and role. A vendor can be limited to the assets it maintains while the control room sees the whole network.
A pump trips in the early hours. By morning, the reservoir it fills is falling. Select any object to see its properties, the objects it links to and the actions it allows.
SCADA reports the trip. Inflow to RES-14 halves to 180 m³/h while morning demand starts to climb.
A predictive rule on RES-14 projects consumption forward and finds the reservoir will reach 15 % at 09:50.
TKT-20931 is created against PS-07, with the asset, its history and its location attached, and assigned to ENG-112, 18 minutes away.
The two-hourly level summary arrives with RES-14 flagged, so nobody has to ask for an update.
Map, graph, table and timeline are four views of the same objects. There is one record underneath, so a level reading, a ticket or a status change shows up in every view at once.
Asset icons change colour on fault or alarm, and update as issues are resolved or escalated. Find the nearest qualified person and dispatch from the map.
Sample dataStart from any object and expand its neighbourhood one hop at a time. Here RES-14 sits at the centre, three links from the engineer heading to the pump that caused the fall.
Sample data| Object | Type | Status | Key values | Linked to | Source |
|---|---|---|---|---|---|
| PS-07 | Pump station | P2 tripped | 180 m³/h · 5.8 bar | RES-14 · TKT-20931 | SCADA |
| RES-14 | Reservoir | Falling | 34 % · 2,720 m³ | PS-07 · LT-14A · PIPE-3318 | PLC · GIS |
| TKT-20931 | Field ticket | Assigned | High · 60 min target | PS-07 · ENG-112 | Workflow |
| ENG-112 | Engineer | En route | 6.2 km · ETA 18 min | TKT-20931 | Mobile GPS |
| LT-14A | Level sensor | Reporting | 2.38 m · every 60 s | RES-14 | PLC |
| PIPE-3318 | Pipe segment | In service | DN400 · 2.4 km · 3.6 bar | RES-14 · WT-22 | GIS · EAM |
| WT-22 | Water tower | Normal | 72 % · 1,500 m³ | PIPE-3318 · DMA-NE-04 | PLC |
| DMA-NE-04 | District metered area | Normal | 610 m³/h · 3.1 bar | WT-22 | SCADA · GIS |
The same objects as rows, ready to filter, sort and export into reports. Every row opens back into the object, with its links and history intact.
Sample dataPS-07 · pump P2 trippedSCADA
Inflow down to 180 m³/hPLC telemetry
2-hourly summary sentMobile · 4 recipients
Forecast: critical at 09:50Low-level rule
TKT-20931 raisedWorkflow · high priority
ENG-112 assigned · ETA 18 minDispatch
Every event on an object and its neighbours, in order, beside the measurement that matters. The forecast runs ahead of the line, which is where the three hours of warning come from.
Illustrative scenario · sample dataThe Classifications Manager holds profiles for people, assets, clients, vendors and tenants, organised with classifications and parent-child relationships. The same structure sets who can see and do what.
Set access once on Storage and every reservoir and water tower beneath it follows. A new reservoir inherits the right properties and permissions from the moment it exists.
A digital twin in Affectli is a schematic view of any asset or group of assets, drawn from the ontology. A schematic is only as useful as the relationships behind it, so every symbol opens the object it stands for.
Draw a single pump, a whole station or the path from source to district, and keep it live from the objects it shows.
Open a symbol to reach the object behind it: its readings, drawings, history and open work, one click from the schematic.
The twin, the map and the table read the same ontology, so a status change appears in all three at once.
Affectli applies AI to objects, so a forecast or an anomaly arrives already attached to its asset, its location and the people responsible. It works on live and historical data, with no lock-in to a single AI vendor.
Models run on data as it arrives, raising real-time alerts and automated responses. The low-level forecast on RES-14 is one: it projects live consumption forward and opens a ticket three hours before the reservoir reaches its critical level.
Models work across historical data to surface trends nobody asked about. Here, ninety days of night flow in one district metered area reveal a slow rise that points to a possible new leak.
Mix, match or upgrade AI capabilities without being locked into a single vendor or platform.
Classify, cluster and detect anomalies across objects and their histories.
Forecast depletion, demand and failure from consumption and condition data.
Work with text held on objects, such as field reports and ticket notes.
Plain answers for teams weighing an ontology against the dashboards, data warehouses and digital twins they already have.
One connected layer across SCADA, PLCs, GIS, CCTV and business systems. Connects to everything you run. Replaces nothing.
See how it connects → Data sovereigntyData ownership by contract, access by classification and role, and a choice of SaaS, PaaS or on-premises.
Data sovereignty & security → Built by Mi-C3Energy, water, transport and telecoms can be modelled as one connected system, so operators and the authorities that supervise them start from the same facts.
The ontology for critical infrastructure at Mi-C3 ↗
A product specialist will model it with you as objects, links and actions, then show how the ontology answers it from the systems you already run.