Mobile GIS Data Collection for Utilities: Field Workflows That Cut Rework

What is mobile GIS data collection for utilities?

Mobile GIS data collection means crews capture asset locations, details and photos on a phone, tablet or rugged handheld. Then the app syncs those records into the utility’s GIS. If the app checks each record in the field, it reaches the office ready to use.

For electric, water and gas networks, the scope is wide. Crews record poles, transformers, valves, meters, manholes and cable routes. They also log condition, defects and trees that grow too close to lines. Because each record feeds planning, outage response and the asset register, a weak record costs money twice. First, you pay for the site visit. Then you pay again to fix the record in the office.

This guide looks at workflow design rather than hardware. Our field data collection services page explains how RedPlanet runs surveys from start to finish. Here, we focus on the design choices that decide whether a crew must go back to site.

Where does rework in utility field data come from?

Most rework comes from three places. First, crews type free text. Next, they save points without a known accuracy. Finally, nobody checks the record before the crew leaves site. Each one is a design flaw, so each one can be designed out.

Free text is the most common cause. For example, one inspector types “wood”, another types “timber”, and a third leaves the field blank. As a result, the asset team cannot filter poles by material without a clean-up job. Likewise, a point from an unknown GNSS fix looks exact on screen. But nobody can tell if it sits within one metre or ten.

Missing photos and blank answers also tend to surface weeks later, in back-office review. By then, the crew has moved to another district. So a five-minute fix on site turns into a half-day revisit. The rest of this guide tackles each cause in turn.

How should you design forms, domains and validation at capture?

Design the form from the data model outwards. First, start with the enterprise GIS schema. Next, apply attribute domains to each field that holds a category. Then build a short form that shows only the questions each asset needs.

Use domains, not free text

Attribute domains are rules stored in the geodatabase. They limit what a field can hold. For example, a coded value domain limits a field to a list, such as pole material. In contrast, a range domain limits a number to a set minimum and maximum, such as pole height. Because the rule lives in the database, every app that edits the layer also follows it.

Show less, calculate more

Modern form tools support conditional visibility and calculated fields. For instance, in ArcGIS Field Maps, a question can appear only when an earlier answer makes it relevant. Calculated expressions can also fill in the inspection date, the user’s name, or the region a point falls in. So crews answer fewer questions and make fewer typing errors.

There is one trap, though. Esri’s own guidance warns that a hidden form element loses its value. So test every branch of the form on a real device before rollout, not just in the desktop designer.

• Required fields: mark only what the asset register truly needs. Otherwise, crews type dummy values just to get past the form.
• Defaults: set a sensible default for the common case, such as asset status “In service”.
• Groups: split long forms into parts, so crews see identity, condition and defects in turn.
• Related records: store repeat inspections in a related table, rather than as new copies of the asset.

Which GNSS accuracy does each utility task need?

Match the receiver to the decision the data supports. Phone GNSS suits condition checks on assets already mapped. However, new assets, buried networks and anything that drives design or digging need a sub-metre or RTK receiver.

The US government’s GPS performance page says GPS phones are typically accurate to within a 4.9 m radius under open sky. That figure gets worse near buildings, bridges and trees. In contrast, high-end users with dual-frequency receivers or correction systems can get real-time positions within a few centimetres.

Malaysia has a real edge here. JUPEM, the Department of Survey and Mapping Malaysia, runs MyRTKnet. It is a national network of always-on reference stations that streams live corrections. Its coordinates use the national datum, GDM2000. So set the right datum transformation in your app. Esri’s high-accuracy data collection guide warns that a missing or wrong transformation makes points sit away from existing data.

Also, keep the metadata. Field Maps can write horizontal accuracy, PDOP, fix type, satellite count and correction age to fields on point layers. Reviewers can then filter out any point saved without a fixed solution.

How does offline GIS sync work when crews lose signal?

Offline GIS lets crews download a set map area and work on the device with no signal. Then they sync their edits when they reconnect. So plan offline areas before go-live, because they decide what crews can see and edit.

The main app families both work this way. In ArcGIS Field Maps, authors set offline map areas in Field Maps Designer, and the layers must have sync turned on. Edits then stay on the device until the worker syncs. Similarly, QFieldCloud packs offline layers into one GeoPackage for QField. Later, it sends back only the changed features and values.

Mobile coverage in Malaysia is better than many crews assume, but it is uneven. Under JENDELA Phase 1, 4G reached 96.92% of populated areas by the end of 2022. However, that figure tracks where people live. Power line routes, plantations, forest reserves and hills can still lose signal. So design every workflow to run fully offline, and treat signal as a bonus.

• Set offline areas by feeder, zone or work package, rather than by whole state.
• Sync at least once a day, so a lost device costs one day of work at most.
• Push all pending edits before you change the schema. Otherwise, field copies fall out of step.
• Skip attachment downloads for crews who only add new assets, so packages stay small.

What photo and attribute standards should field crews follow?

Write a short capture standard and build it into the form. Each photo needs a clear purpose, the same framing and a link to one asset. That way, reviewers can check details without a site visit.

Photos do most of the work in QA. For example, a reviewer can confirm a pole’s material, cross-arm type and tag number from one good picture. But a blurry close-up of rust proves very little. Tropical weather adds its own problems, because rain, glare and damp on the lens all spoil images. So the standard should tell crews when to retake a shot.

Finally, keep field names and units the same as the enterprise schema. If they differ, every sync needs a mapping step, and mapping steps are where values go missing.

How do you run QA at source and load data into enterprise GIS?

Run QA in three layers. First, set rules on the device. Next, add a supervisor review each day. Then run automated checks before data enters the enterprise GIS. Each layer catches what the last one missed, and errors cost less to fix the earlier you find them.

1. On the device: domains, required fields and accuracy limits block the most common errors as crews capture.
2. Daily review: a supervisor checks a sample of records, photos and GNSS data. Then failures go back to the same crew while they are still nearby.
3. Before loading: automated checks test topology, links and attributes against the network model. After that, a controlled load moves data into the system of record.

That last step matters most for utilities, because the target is often a connected network model, not a flat layer. For example, a new service cable must link to the right transformer, and a valve must sit on a main. Our GIS QA/QC services apply these rules before field data reaches platforms such as GE Vernova Smallworld or Esri. Where old records also need cleaning, we pair this with GIS data migration.

Which mobile GIS apps suit utility asset inspection work?

Choose the app that fits your enterprise GIS and your workflow, not the longest feature list. Most utilities pick from three groups. These are Esri apps, the open source QGIS family, or a utility platform with inspection and outage tools built in.

What does a realistic mobile GIS rollout look like?

A sound rollout moves from schema to pilot to scale. It also runs a field pilot before any wide launch. The steps below show the order, while the real timeline depends on asset classes, crew numbers and integration scope.

Along the way, the same mistakes show up on many projects. However, most are cheap to avoid if you spot them early.

Frequently Asked Questions

How do utilities cut rework in mobile GIS data collection?

Utilities cut rework by building quality into the capture step. Domains replace free text, and GNSS metadata proves accuracy. In addition, offline packages keep crews working, while daily review catches errors near the asset. The app matters, but the workflow matters more.

If your network is electric, start with the assets that drive outage and planning choices, such as poles, transformers and service links. Our work on GIS for electric utilities shows how clean field data feeds the network model. So when you are ready to review your own workflow, get in touch with our GIS consultants.