GIS for water utilities means holding every pipe, valve, hydrant, reservoir, pump station, meter and service connection in a geographic information system (GIS), with its attributes and network connections. A good inventory answers not "where was it?" but "who loses supply if we shut this valve?"; hydraulic models and loss analysis draw on it.
Key takeaways
- A water network GIS inventory holds pipes, valves, hydrants, reservoirs, meters and service connections with their attributes and network connectivity.
- Türkiye's Ministry of Agriculture and Forestry puts the 2024 drinking water loss rate at 31.6%; large municipalities must reach 25% by 2028.
- The inventory is built in eight steps: data model, sources, digitising, topology, confidence classes, field checks, customer matching, update workflow.
- Giving every record a confidence class such as "surveyed", "from drawings" or "estimated" lets uncertain data be used without being deleted.
- Unless the crew doing a repair records the change in a mobile app, the inventory falls behind the network within a few years.
This guide is not about rolling out GIS across a whole council. It covers one thing: the water network inventory itself, meaning which assets, which attributes, what level of accuracy and how to keep it current. For the wider municipal picture, see our guide to GIS for municipalities.
Why are water network records always incomplete?
In most water utilities, network data has built up in layers. Older districts exist only as scanned paper sheets, newer developments sit in a contractor's CAD drawings, and recent repairs are recorded only on job cards. The same main can appear with three different diameters in three sources, and only the field crew knows which one is right.
The real gap is usually not the line on the map but the attributes and the connectivity. A pipe being drawn does not mean anyone knows when it was laid, what it is made of or which valves isolate it. Service connections are often missing altogether; the customer exists only as an address in the billing system.
An inventory bought as a one-off survey is most accurate on handover day; every repair and new connection then ages it. Without a way for crews to record changes, they soon go back to trusting experience over the map.
What does a weak water network inventory cost? Loss targets and wasted time in the field
In Türkiye, water utilities (usually municipal water and sewerage administrations) work to a statutory timetable, and a reliable network inventory is a precondition for the measures needed to meet it.
The Ministry of Agriculture and Forestry's Water Efficiency portal, in figures published as of 2026, reports that the average loss rate in Türkiye's drinking water networks fell from 39% in 2015 to 31.6% in 2024; metropolitan and provincial municipalities must cut their losses to at most 25% by 2028. According to the ministry, loss rates in developed countries range from 8% to 24%, and the ultimate target is not 25% but around 10%.
For inventory work, the more telling part of the Water Efficiency portal is its list of measures: it explicitly includes building a geographic information system infrastructure, using hydraulic models and setting up district metering and pressure management areas. None of these works without an accurate network inventory, because a hydraulic model is built from the diameter, material and connectivity held in it. Here, GIS is not a visualisation tool but the database the measures are calculated on.
District and town municipalities are covered too. The ministry's announcement on the water loss work schedule explains that Türkiye's 2014 Regulation on the Control of Water Losses in Drinking Water Systems caps losses for these municipalities at 30% by 2028 and 25% by 2033. We cover what these targets mean at council level in our municipal GIS guide; here we focus on the network inventory itself.
On the ground, a weak network inventory means crews hunting for valves and digging in the wrong place. Methods for measuring loss are in our water loss analysis guide; the network inventory is the map that analysis runs on.
GIS for water utilities: which assets go into the inventory?
Define the scope of the water network inventory in writing before any digitising starts. The table below lists the core assets of a typical drinking water network and the minimum attributes each should carry.
| Asset | Minimum attributes | Why it matters | Typical source |
|---|---|---|---|
| Pipe (trunk, distribution) | Diameter, material, year laid, pressure class, depth | Hydraulic model, renewal priority | Design CAD, repair records |
| Valve | Type, diameter, normal position (open/closed), last operated | Isolation, zone boundaries | Field survey, operations logs |
| Hydrant | Type, connected main, last inspection | Fire cover, flushing | Fire service and operations lists |
| Reservoir and pumping station | Capacity, pump details, connected main | Supply zones, energy use | SCADA, plant drawings |
| Zone flow meter and pressure point | Location, boundary valves, communications | District metered areas (DMAs) | Operations, SCADA |
| Service connection and meter | Connected main, meter number, customer reference | Affected customers, water balance | Billing, meter reading |
In the asset table, the relationship fields such as "connected main" and "boundary valves" matter most. Features drawn without being joined make a tidy map that does not behave as a network, so no trace or isolation query gives a correct answer.
Linking customer references and addresses means processing personal data under Türkiye's Personal Data Protection Law (KVKK, broadly comparable to the GDPR). Decide early which roles may see customer details; our KVKK compliance steps summarise the obligations.
How do you build a water network GIS inventory? An 8-step method
- Write the data model. Define asset types, mandatory attributes, value lists (such as material codes) and relationships in one document. The same document becomes the delivery requirement for contractors; our article on writing technical specifications shows how to phrase it.
- Gather every existing source. Design drawings, scanned sheets, job cards, the billing extract and the SCADA (supervisory control and data acquisition) point list go into one workspace. Nothing is discarded; each record keeps a source tag.
- Digitise and unify coordinates. Scanned sheets are georeferenced (fitted to real-world coordinates), and data in legacy systems such as ED50 is transformed to Türkiye's ITRF96-based national coordinate system. Orthophotos or up-to-date imagery from drone data processing provide a backdrop for checking against roads and parcel boundaries.
- Apply topology rules. Automated checks list pipes whose ends do not meet, valves that sit off the line, meters attached to nothing and self-intersecting mains.
- Give every record a confidence level. Keep a class such as "surveyed", "from drawings" or "estimated". Uncertain data is not deleted, but everyone can see how far to trust it.
- Verify in the field. In low-confidence areas, crews locate valve covers, fix positions with a GNSS (global navigation satellite system) receiver and use pipe locators where needed. Each verified record moves up a class.
- Match service connections. Customer references and meter numbers from billing are linked to the main that supplies them. Where one customer has several records, this is the moment for duplicate record clean-up.
- Switch on the update workflow. Repairs, new connections and valve changes are recorded by the crew doing the work in a mobile app, then approved centrally before they reach the inventory.
Inventory quality checklist
Running these checks on the water network inventory at handover and at the end of each period shows whether the data is reliable enough for decisions. Each utility sets its own acceptance thresholds.
| Check | How it is measured | What a failure means |
|---|---|---|
| Attribute completeness | Share of pipes and valves with empty mandatory fields | Hydraulic model and renewal plan built on gaps |
| Topology errors | Count of unconnected ends and off-line valves | Isolation queries return the wrong customer list |
| Field-verified share | Share of records classed as "surveyed" | Risk of excavating in the wrong place |
| Customer matching | Meters linked to a main vs. total active customers | Zone water balance cannot be calculated |
| Update lag | Time between repair date and inventory update | The map falls behind the network |
Inventory quality checks are really a data governance task: unless someone owns each field and each correction, the table shows the same problems every period. We cover the framework in our data governance guide and keeping a single trusted source of asset codes in our master data management guide.
What does a connected inventory make possible? Hydraulic models, DMAs and valve isolation
Once assets are properly joined, a water network GIS inventory supports work that often looks like separate projects:
- Valve isolation and affected customers. Select a burst location and the network calculates which valves to close and which customers lose supply.
- DMA design. Zone boundaries are defined by boundary valves and flow meters, and customers inside each zone are grouped automatically.
- Hydraulic modelling. Diameter, material and connectivity feed the model, where pressure management scenarios can be tested.
- Renewal priorities. Year laid, material and burst history are combined to decide which mains to replace first.
- Live data on the map. Readings from reservoirs and pumping stations arrive through SCADA systems, and customer consumption through automatic meter reading, attached to the right feature.
How the pumping side is instrumented is covered in our guide to remote pump station monitoring. To track losses, bursts and repair times by zone on one screen, we recommend choosing indicators with the method in our BI dashboard and KPI guide.
How we build water network inventories at Digital Bridge
Digital Bridge is a technology company that has developed software and hardware in Adana since 2013. We do not sell an off-the-shelf GIS licence; we build the inventory around your asset types, workflows and permission structure. A typical engagement runs in this order:
- Discovery and data audit. We review your drawings, sheets, repair records and billing data and write the data model with you. Scope, phases and fees are set out in a written proposal.
- Pilot zone. We digitise, run topology checks and verify in the field for a single supply zone, then review the quality checklist together. Only once the method works there do we move to other zones.
- Web and mobile GIS. We load the inventory into a web-based geographic information system and give crews an app that works offline for updates. Repair work orders link to the same records through field service management; the general logic is explained in our field service management guide.
- Integration. We connect the inventory to billing, to automatic meter reading (AMR) and to SCADA. Our approach for water, power and gas networks is on the energy and utilities page, and the municipal scope on the public sector and municipalities page.
- Data quality. We turn the checklist into automated reports and define owners and correction workflows through our data governance and quality work.
Because the same team develops our meter reading (OSOS) devices, their embedded firmware and the web and mobile software, the link between map and field devices stays under one responsibility. We work across all of Türkiye, remotely and on site. For related topics, browse all our Data & Analytics guides.
Next step
Before starting a water network inventory, answer three questions about your own network: what share of pipes have diameter and material recorded, what share of valves have been verified in the field, and what share of customers are linked to a main? The answers tell you where the pilot zone should be. Then get in touch with us, and we will review your data together and draft the data model and pilot plan.