Digital Twin as a product and three operational areas where it has an immediate impact on decision-making — water losses, sewer systems and WWTPs.
Core product.
For us, an operational digital twin is a “live” model of a network or facility that is synchronised with operational data and ready for scenarios. The foundation is integrating the model with GIS and SCADA/telemetry so it can be updated and validated continuously.
In practice this makes it possible to identify deviations quickly, understand system behaviour and compare “what should be” with “what is”. The operator gains context for the data — not just charts, but the impact across the whole area and on critical nodes.
The decision-support layer rests on the principle that a digital twin must answer concrete questions: “what happens if…”, “where does a measure deliver the most”, “how should operations change during a failure”. We build scenarios for interventions, regime changes, development, extremes and failures, and express the results as KPIs and map outputs.
The output is therefore ready for a technical decision, not just a “modeller’s report”. In practice this shortens incident response times and improves the quality of investment planning.
For utility teams there is great value in practising responses to failures or non-standard states without risk to real operations. Within a digital twin solution we can prepare training “what-if” scenarios that simulate typical incidents and show their impact across the network.
This turns the digital twin into a tool for know-how transfer and readiness — less improvisation and faster, safer interventions in the field.
We build water loss management on a combination of data (flows, pressures, consumption) and a hydraulic model that can determine where the system is pressure-sensitive and where the savings potential is greatest. The basis is NRW diagnostics in spatial context and the identification of areas where pressure reduction delivers a real effect without degrading service.
The model allows alternatives (PRV settings, pumping regimes, interconnections) to be compared before any intervention. The output is a set of proposed measures with quantified impact and recommended priorities.
In larger systems it makes sense to establish DMAs (district metered areas), which allow more accurate balance monitoring and faster problem localisation. We help design DMA boundaries based on network topology, operational dependencies and the available metering infrastructure.
A cost-benefit assessment is included: where a DMA has the best benefit-to-cost ratio and what equipment is required. In the digital model we can verify the hydraulic consequences of closing interconnections and the effects on redundancy.
When a leak is suspected, narrowing the search to the most likely reaches is critical to save field teams’ time and cost. We use a combination of pressure and flow data, regime changes and model simulations to estimate in which branches a leak would best explain the observed deviations.
This is not “magic AI” but engineering-grounded shortlisting of candidates that a field team can verify. The result is a smaller search area, better planning and a higher detection success rate.
We prepare CSO solutions as a combination of model analysis, design of measures and comparison of alternatives by effectiveness and cost. We focus on identifying the network “bottlenecks” that cause overflows and local flooding, and on proposing the steps with the greatest effect.
The model allows testing of detention structures, route changes, pumping modifications or other operational strategies. The outputs are prepared to be usable for investment planning while remaining defensible for compliance and stakeholder communication.
I/I tends to be a hidden problem that significantly increases inflow to the WWTP, worsens CSOs and raises treatment costs. We diagnose it by combining measurement and rainfall analysis with model-based decomposition of flows into a base component and rainfall contributions.
The aim is to identify the areas where rehabilitation delivers most and to set intervention priorities — work is done where the effect is highest.
For operational control it is essential to know what happens under different types of rainfall and how to prepare operations to minimise the impact. In digital twin mode we can prepare scenarios and simple decision rules for setting structures (pumping, overflows, detention) according to the current state of the system.
The result is a procedure usable without the operator having to “operate the model”. It is a practical form of decision support for sewer systems.
A process digital twin of a WWTP is a virtual model of the treatment lines, tied to operational data and used for optimisation and prediction. It allows setting changes to be tested, load fluctuations to be handled and treatment stability to be improved without risk to real operations.
It also matters in practice that the digital twin can serve as a platform for planning and operator training, especially for complex processes.
Energy optimisation at a WWTP typically concerns aeration, pumping, mixing and chemical dosing. Based on the process model and data, we can propose setpoints and regimes that reduce energy consumption while maintaining effluent quality.
The advantage of the model is that it allows operating alternatives to be compared and the impact estimated before any change to the technology. Used long term, it allows continuous assessment of whether measures are working and where further improvement is possible.
Predictive maintenance means connecting operational data, model context and asset records so that at-risk elements are identified before they fail. In process facilities this often comes down to being able to quickly find a specific asset, its history and its role in the system.
Combining a 3D/asset layer with data and maintenance records makes it possible to build practical workflows for maintenance and intervention planning.
