Norwegian fjord, InfoTiles Delta AI assistant for water and wastewater network management
AI Decision Support · InfoTiles
Δ Delta

Delta: AI Decision Support for Water and Wastewater Utilities

Most water networks hold decades of operational knowledge. It lives in databases no one queries. Delta Delta makes it answerable in plain language, in seconds.

Book a demo See Delta running against your network data.
Water Network Decision Support

AI Decision Support for Water and Wastewater Network Management

Delta Delta is an AI assistant built on top of your water or wastewater network. It translates the structural knowledge locked inside that model (topology, flow estimates, risk scores and telemetry) into direct, conversational answers.

Ask about a pressure zone, a failing asset, a catchment with elevated infiltration, or the consequence of a planned intervention. Delta Delta returns a reasoned answer grounded in your actual network data, not generic hydraulic knowledge.

No calibrated hydraulic model required. Delta Delta delivers flow estimates and fault context using graph algorithms across pipe topology, removing the single biggest barrier to AI adoption in water utilities.

"The LLM is the reasoning layer. The network model is the knowledge base. Neither is sufficient without the other."

Data Sources: SCADA, Risk Models & Network Analytics

Telemetry, Risk Scores, and Graph Based Hydraulics: Built on Data You Already Hold

Answers are composed from data layers that PipeFusion already holds, so no new data collection is required. The reasoning layer sits on top of existing water utility asset management models.

Data prerequisites. PipeFusion network wide data preparation is the supported starting point. The minimum inputs are the network structure as GIS data or a graph based hydraulic model, SCADA telemetry for operational questions, and rehabilitation cost data for cost ranked results.

Layer 1
Live telemetry (SCADA)Read only access to pressure, flow, alarms and operational events. Delta never writes to or controls SCADA.
Layer 2
Graph based network modelHydraulic parameter estimation using graph algorithms across the pipe topology.
Layer 3
PipeFusion risk scoresPer asset condition index, failure probability, and remaining service life estimates.
Layer 4
Rehabilitation cost dataUnit costs for repair, lining, and replacement tied to pipe material, diameter, and condition.
Capabilities: Pipe Rehabilitation, I&I Reduction & Risk Analysis

Risk Analysis, Rehabilitation Prioritisation, and Inflow & Infiltration Analysis

Delta Delta's scope is deliberately bounded to areas where InfoTiles data provides genuine depth. Capabilities outside this scope are not surfaced.

Pipe network risk analysis

Query risk scores, explain the drivers behind a specific segment's ranking, and compare candidate assets for intervention sequencing.

Rehabilitation prioritisation

Generate cost ranked intervention lists from risk data and rehabilitation method costs, with transparent methodology.

Fault context on demand

Surface hydraulic context around an active alarm, including upstream topology, adjacent risk scores and recent telemetry, without manual correlation.

Reduce inflow and infiltration (I&I)

Combine SewerIntelligence I&I telemetry with PipeFusion risk scores to prioritise CCTV inspection plans, structured and ranked.

PipeFusion risk model SewerIntelligence I&I Graph based hydraulics Rehabilitation cost estimation Water utility asset management

Measurable outcomes. Success is measured against a baseline taken before deployment: engineering time per analysis, quality of prioritisation, reduction in I&I, and consistency of decisions.

AI for Wastewater Network Optimisation

Decision Support in Action: I&I Inspection Prioritisation Using Live Network Data

The example below is an actual prompt run against a utility network. Delta Delta combines SewerIntelligence I&I telemetry with PipeFusion risk scores to generate a prioritised CCTV inspection plan that is structured, ranked and ready to act on.

Delta AI assistant: CCTV I&I inspection prioritisation using PipeFusion risk scores
Live query · SewerIntelligence I&I telemetry + PipeFusion LoF/CoF risk scores
Water Network Operator, Utility on Delta

What Water and Wastewater Utility Teams Say

"We went through the solution with our IT department, and they were impressed by how InfoTiles handled data security. Everything runs within Microsoft, no data leaves the system, and nothing is used for model training."

Network Manager, Utility

"The AI assistant does the same calculations that would take me two days and digging around in the database to find."

Operations Engineer, Utility

"If you are wondering about something, you can actually ask follow up questions."

Network Operator, Utility

"It will find the same thing I find, just much faster."

Utilities Analyst, Utility
Architecture: RAG, Azure AI Foundry & Human in the Loop

How Delta Works: Retrieval Augmented Generation on Your Network Model

Delta Delta uses a retrieval augmented generation (RAG) approach. At query time, relevant segments from the network model, risk scores, and reference documentation are retrieved and injected into the LLM context. The model does not retain state between sessions and is not fine tuned on customer data.

Infrastructure: Microsoft Azure AI Foundry, EU hosted, stateless inference, no training on customer data. The LLM is replaceable; the retrieval and context construction pipeline is where the domain specificity lives.

A person approves every output. Delta Delta cannot make decisions autonomously. All outputs require review and approval by a person, and Delta supports decisions without taking them.

Network Decisions, Not Office Work

Delta Compared with ChatGPT and Copilot

ChatGPT and Microsoft Copilot are general assistants designed for text, documents and everyday office work. Delta Delta is a decision support assistant designed for decisions about the network itself. PipeFusion represents your network as a graph neural network, so Delta can reason about pipes, flows and connections rather than words alone, and every answer is rooted in your own operational data.

Many utilities will use a general assistant for office productivity and Delta for decisions about the network. The two complement each other.

Security, Compliance & Data Governance

EU Hosted, Designed for NIS2, and Built for Critical Infrastructure Procurement

Delta Delta is Norwegian built and operated. For utilities managing critical infrastructure, this matters structurally.

EU hosted infrastructure

Runs on Microsoft Azure AI Foundry within EU/EEA regions, under standard EU/EEA data processing agreements.

Stateless inference

Each session is independent. Nothing persists beyond the session boundary. No customer data is used for model training.

Designed to align with NIS2

Delta is designed to align with NIS2 obligations and Norwegian security principles for cloud services. Compliance documentation can be reviewed as part of a procurement process.

Controlled access

Access is governed through Microsoft Entra ID with role based access control, multi factor authentication and audit controls. External guest access is supported, for example for consultants working with a utility.

Logged queries and responses

Queries and responses are logged to support auditability and compliance, and the audit logs can be reviewed by the customer.

Commercial & Deployment

Pricing with a Cost Ceiling, Operational within Weeks

Delta Delta runs on an existing InfoTiles deployment and is priced so that the customer stays in control of cost.

Two billing options

Billing on tokens used, or at a fixed monthly rate. Under the fixed rate, Delta is switched off for the rest of the month if the cost reaches that rate, so the monthly cost cannot exceed it.

No new infrastructure

An existing InfoTiles deployment and outbound HTTPS connectivity are all that is needed. No new infrastructure and no new sensors.

Operational within weeks

Delta is typically operational within weeks once integration is complete.

Minimal onboarding

Onboarding focuses on the data and on interpreting outputs. Users work in natural language, so there is no new tool syntax to learn.

Customer references and case studies are available on request.

Why Delta

Why Delta: The Symbol of Change in Water Networks

Δ mark
The symbol of change

Δ is the mathematical symbol for change: the measured difference between one state and the next. In water networks, that difference is the signal: a pressure anomaly, a flow deviation, a demand shift. Delta surfaces it in plain language, in seconds.

Delta The name carries additional resonance. In hydrology, a delta is where flows converge and are understood, where a network meets the wider world. And like the Norse varder, the stone cairns built on high ground so those who follow always know where they stand, Delta assembles accumulated knowledge and passes it forward.

Deployment & Availability

Delta Is Live with Norwegian Water Utilities

Delta Delta is now in production with Norwegian water utilities as the first wave. Expansion to further markets is underway.

See Delta Delta running against your network data.

Book a demo

Frequently asked questions

Trust and explainability

Will Delta give the same answer when the same question is asked again?

The substance is consistent: like the opera houses in Sydney and Oslo, answers may look different on the surface while resting on the same engineering principles. Methods and data queries are fixed, so the same question on the same data rests on the same underlying results, whoever asks. A precise question gives a more robust answer than a vague one.

What happens when Delta cannot answer a question confidently?

Delta is explicitly instructed not to invent answers. Where information is missing or uncertain, the answer states what could not be established and why, rather than filling the gap with a guess.

How does Delta explain its recommendations, and can we see the reasoning?

Yes. Delta exposes the inputs, risk scores, network parameters and cost assumptions behind each recommendation, so every step from data to conclusion can be reviewed.

When does human review remain necessary?

Always. Delta cannot make decisions autonomously, and all outputs require review and approval by a person.

Why not use ChatGPT or Microsoft Copilot?

They solve different problems. ChatGPT and Copilot are general assistants for text, documents and office work, while Delta is built for decisions about the network, using the PipeFusion graph representation and your own operational data. Many utilities will use both, and the two complement each other.

Can Delta replace my SCADA or asset management system?

No. Delta is a decision support layer that surfaces evidence and ranked recommendations at the point a decision is needed. Approval remains with the operator.

Data and integration

What data and setup does Delta need?

PipeFusion network wide data preparation is the supported starting point. The minimum inputs are the network structure as GIS data or a graph based hydraulic model, SCADA telemetry for operational questions, and rehabilitation cost data for cost ranked results. No new data collection is required beyond data the utility already holds.

Is Delta integrated with SCADA?

Yes, in one direction only. Delta has read only access to live SCADA data such as pressure, flow, alarms and events, and it never writes to or controls SCADA.

Does Delta need a calibrated hydraulic model?

No. Delta uses graph based parameter estimation derived from the network structure, so deployment does not depend on building or calibrating a hydraulic model first.

Can Delta analyse both water and wastewater networks?

Yes. Delta works across WaterIntelligence, SewerIntelligence and PipeFusion contexts and scopes each query to the relevant network automatically.

What happens if network data or telemetry has gaps?

Delta does not hide data quality issues. It flags offline sensors and anomalous values, surfaces uncertainty, and traces each recommendation to the evidence that is available. Live and historical data are treated separately, so a sensor outage is visible instead of silently affecting results.

Use cases and value

Is Delta for project, operational or strategic decisions?

All three. A project team can ask for an I&I remediation plan for the coming quarter, an operator can investigate a SCADA alarm with upstream topology and recent telemetry in one response, and an asset manager can compare cost ranked rehabilitation lists under different cost assumptions.

How does Delta help prioritise pipe rehabilitation?

Delta combines PipeFusion risk scores with rehabilitation cost data to produce cost ranked intervention lists. Scenarios can be compared under different cost assumptions, and the methodology and audit trail are transparent.

Can Delta reduce inflow and infiltration (I&I)?

Yes. Delta combines SewerIntelligence telemetry with PipeFusion risk scores to prioritise CCTV inspection and remediation plans, which supports a systematic approach to I&I reduction.

How are the benefits measured?

Success is measured against a baseline taken before deployment: engineering time per analysis, quality of prioritisation, reduction in I&I, and consistency of decisions. These can be set up as a KPI dashboard and form the basis of the business case.

Security and governance

Where is Delta hosted, and who can see customer data?

Delta is hosted in Microsoft Azure AI Foundry in EU and EEA regions. Customer data is not stored by the model service and is never used for model training, and sessions are stateless with no memory between them.

How is access governed?

Through Microsoft Entra ID with role based access control, multi factor authentication and audit controls. External guest access is supported, for example for consultants working with a utility.

Are queries and responses logged?

Yes. Logging supports auditability and compliance, and audit logs can be reviewed by the customer.

Is Delta NIS2 compliant?

Delta is designed to align with NIS2 obligations and Norwegian security principles for cloud services. Compliance documentation can be reviewed as part of a procurement process.

Commercial and deployment

What does Delta cost?

Delta can be billed on tokens used, or at a fixed monthly rate. Under the fixed rate, Delta is switched off for the rest of the month if the cost reaches that rate, so the monthly cost cannot exceed it. An illustrative ROI calculation can be prepared with a utility.

What infrastructure is required?

An existing InfoTiles deployment and outbound HTTPS connectivity. No new infrastructure and no new sensors are needed.

How quickly can Delta be deployed?

Delta is typically operational within weeks once integration is complete.

What training is required?

Onboarding is minimal and focuses on the data and on interpreting outputs. Users work in natural language, so there is no new tool syntax to learn.

Is Delta production ready?

Yes. Delta is a production ready decision support platform, and customer references and case studies are available on request.