Municipal utilities face a twofold challenge. On the one hand, regulatory requirements force them to capture, document and evaluate ever more data. On the other, they are expected to rebuild core infrastructure for the energy and heating transition within a short time, and to do so with ever tighter financial room for manoeuvre.
550 million euros. That is the financing requirement for building predominantly renewable heat networks over the next 15 years. And since we are talking here about the city of Konstanz alone, the scale of the effort ahead for cities, districts and municipalities across the country becomes clear. Few look at the opportunities as optimistically as Konstanz's mayor Uli Burchardt:
“Heat networks are the business model of the future in heat supply. So we are not simply spending money on climate protection – we are investing in our company and in a business model our children will earn money with.”
It is encouraging that politics on Lake Constance is not only chasing short-term political success but thinking long term. Yet these investments also have to be viable in day-to-day operations. The question is therefore not whether investment is needed, but how municipal utilities can meet economic viability, security of supply and regulatory requirements at the same time. Three concepts play a decisive role here: predictive maintenance, vertical data integration and an integrated data platform.
A flood of data meets fragmented systems
Municipal operations are already highly data-driven today, but rarely use that fact strategically. The German Metering Point Operation Act, energy monitoring and reporting duties towards supervisory authorities and funding bodies all require the continuous capture of measurement, operating and condition data. At the same time, the energy transition and the shift towards sustainable mobility are increasing pressure on European electricity distribution grids, driven by considerably greater load fluctuations and rising electricity demand. This can have critical effects on equipment in the distribution grid and reinforces the need to plan, carry out and steer the maintenance of assets.
Predictive maintenance strategies that analyse current and historical condition data of equipment offer promising solutions here. This is not only about designing and optimising analytical algorithms or data spaces. What matters is aligning IT and service delivery closely, because in practice a maintenance service is provided jointly by several parties and supported by IT. A Fraunhofer analysis confirms this.
In practice, however, this data usually sits in separate systems. Grid operations, generation, sales, water and heat use different IT solutions, often grown historically. Added to that is data from generation plants, grids, heat storage, water and wastewater systems as well as from sales and billing. The result is data silos, high manual coordination effort and limited transparency about the actual condition of the infrastructure. Especially at a time of skills shortages, this complexity becomes a risk: specialised data engineers or data scientists are rarely available in municipal operations, while the effort for data maintenance and analysis keeps rising.
Maintenance as a cost factor, predictive maintenance as a lever
One particularly large cost block in municipal operations is the maintenance of plants and networks. Pipe bursts in heat networks and unplanned outages of combined heat and power units, pumps or transfer stations cause not only repair costs but also follow-up costs from supply interruptions, replacement procurement and increased staff deployment.
Many municipal utilities still work with reactive maintenance, meaning repair after failure, or interval-based maintenance, meaning inspection on a fixed schedule. Both approaches are either expensive or wasteful. The more efficient approach is predictive maintenance: assets are maintained when data indicates imminent wear, neither too early nor too late.
What predictive maintenance means in concrete terms: sensor data from a centrifugal pump in the heat network, such as vibration frequency, temperature and pressure profile, is evaluated continuously. If the values deviate systematically from the normal profile, the system issues a maintenance alert early. The bearing is replaced as planned in the next maintenance window, not as an emergency call-out at 2 a.m. with a supply interruption for 400 households.
Experience from practice shows that unplanned outages can be reduced significantly through data-based maintenance. At the same time, the service life of critical components is extended, which lowers tied-up capital and makes investments more predictable. Every avoided emergency call-out relieves not only the budget but also the already scarce personnel resources.
Vertical data integration as the operational foundation
Predictive maintenance only works if the right data is available at the right time in the right place. This is exactly where vertical data integration comes in.
Vertical data integration means bringing data together along the entire operational hierarchy: from the sensor level (field device, control unit) through the process level (SCADA, control technology) to the enterprise level (ERP, billing, reporting). In contrast to horizontal integration, which links systems on the same level, this creates a continuous situational picture across all layers of operations.
For municipal utilities this means, concretely: sensor data from the heat network, maintenance histories from the CMMS, load data from the grid control system and environmental data from external sources are combined in a single, unified data model. Only then do robust forecasts emerge that make maintenance measures plannable and integrable into existing operating routines.
Data platform: the technical prerequisite for both
The key to these effects lies not in individual algorithms but in a stable data platform as the central foundation. A municipal data platform creates the basis for consolidating data from different systems consistently, historising it and evaluating it securely. A capable data platform for utilities fulfils several functions at once:
- Operational control: operating data is available in real time for decisions.
- Predictive maintenance: anomaly detection and forecasting based on historical and current data.
- Critical infrastructure security: clear access rights, transparent data governance and protection of system-critical infrastructure data.
- Regulatory compliance: reporting duties towards supervisory authorities and funding bodies are met efficiently.
Energy and heat networks are part of critical infrastructure. Protecting operational data, clear access rights and transparent data governance structures are prerequisites for implementing digitalisation responsibly. Modern data platforms combine technical security with organisational clarity and build trust with supervisory authorities, partners and within the organisation itself.
In closing: data analysis and predictive maintenance as a way out of the municipal data dilemma
For municipal utilities, the way out of the tension between investment pressure and tight budgets does not lie in new funding programmes or additional reporting duties alone. What is decisive is operating existing infrastructure more intelligently.
Vertical data integration, combined with predictive maintenance on a central data platform, makes it possible to cut costs, reduce risks and prioritise investments better. The energy transition is implemented locally, but its price is also decided in cyberspace: utilities, grid operators and platform providers have to create data spaces in which data flows can be exchanged securely, in standardised form and in a trustworthy way. Those who invest today in clean data architectures and integrated data platforms lay the foundation for mastering this task economically, securely and successfully over the long term.
About control-f. control-f GmbH is a values-driven AI company based in Konstanz. Since 2022, the data boutique has been building big data platforms for industrial telemetry data and helping companies in German-speaking Europe make complex data landscapes usable. Its clients include large corporations and mid-sized companies from plant engineering, automotive and the energy industry. Managing directors Simon Deussen (machine learning engineer and founder) and Daniel Tremer (formerly Specialist Data Science & AI Projects at Porsche AG) focus on building stable data architectures as the foundation for analytics, software solutions and AI applications such as predictive maintenance.
Kategorie / Category: Blogposts