Cover01 / 14

Showcase · 2026

Operational visibility in real time.

How field data, cloud and analytics turn into traceability, visibility and action in a global, high-stakes operation.

14 chapters · ~18 min

Scroll

The challenge

A global operation, complex and high-risk.

A Fortune 500 upstream energy company and a global leader in oilfield services: well construction, completions, stimulation, subsea and production optimization, onshore and offshore, down to deepwater. Remote environments, intermittent connectivity, thousands of assets on the move and tens of thousands of people keeping the operation running 24/7.

120+

Countries

$36B

Annual revenue

200k+

Assets tracked

10k+

People in the operation

24/7

Continuous operation

US · CA · AR · MX · BR · DZ · SA · AE · RU · CN · AU · ZA

The operational context

A continuous ecosystem that runs 24/7.

Every fracking site is one integrated operation: people, materials and information in sync around the clock to keep production, safety and cost under control.

Fracking site: drilling, materials, crews in the field and the remote back office

01Crews

Field crews run the operation on site, in shifts, 24 hours a day.

Technicians and operators
Equipment maintenance
Field monitoring
Safety and quality

02Materials

A continuous flow of raw material arriving and being consumed.

Sand (proppant)
Chemicals
Water
Fuel and supplies

03Back office

Support teams keeping resources available, the operation efficient and compliance on track.

Delivery planning
Operation monitoring
Analysis and alerts
Exception management

Where it hurt

Three classes of problem. One root cause.

The data existed. It just wasn't reaching the right place, at the right time, in the right shape, for the people who had to decide.

Asset visibility and tracking

Hundreds of heavy assets: pumps, blenders, sand silos, water tanks, frac vans, moving in, out and between sites. A different mix of systems and providers in every region, with no unified view of where anything is.

4 systems10+ GPS providers3-day audit

Supply delivery logistics

Sand, water and chemicals arriving in a continuous flow. If the sand doesn't land on time, production stops. Dispatch ran on spreadsheets and phone calls: no real-time tracking, no auto-matching, no demurrage visibility.

Sand-out = downtimeManual dispatchExcess inventory

Field activity logging

Crews logged stages, pressures and critical events in spreadsheets, out in the field, wearing gloves, with no signal. Data that feeds customer billing, compliance and the coordination between field and control center.

Stage eventsCustomer billingSync hours later

“The three problems shared one root cause: the operation needed an orchestration layer that could connect data, decision and action.”

The solution, in one sentence

A digital layer over the existing systems that captures, integrates, organizes and activates operational data.

captureintegrationnormalizationgovernanceanalyticsMLsmart appsalerts

An architecture that connects field apps, legacy systems, sensors and events into a single, reliable source of operational truth, running on top of the systems already in place.

Layer 01

Every asset in a single view.

Asset visibility and tracking

A unified asset map with RFID, QR Code and GPS tracking. Result: inventory audits done in 15 minutes, with no manual consolidation.

RFIDQR CodeGPSGeofencing

BigQuery · Cloud Storage · Pub/Sub

Layer 02

Dispatch had outgrown the spreadsheet.

Supply delivery logistics

Real-time tracking, geofencing per well and automatic vehicle-to-demand matching. Result: 100% of deliveries tracked.

Real-timeGeofencingAuto-matchingETAs

Pub/Sub · Cloud Run · Dataflow

Layer 03

Data that starts where the work happens.

Field activity console: job session, event log and stage detail

Field activity capture

Real-time capture in the field, on an offline-first app with automatic sync. Result: 5× faster sync and 100% data consistency.

Offline-firstAuto-syncPhoto + GPSVoice notes

Cloud Functions · Storage · Vertex AI

Solution architecture

From field to decision.

01Sources

Field · sensors · systems

Field apps · GPS / RFID / QR · ERP / TMS / CRM · IoT sensors

02IngestionPub/Sub

03Data PlatformBigQuery

04IntelligenceVertex AI

05Action

Dashboards · apps · alerts

Executive dashboards · Smart apps · Real-time alerts · Mobile UIs

Google Cloud provides the foundation of scale, data and intelligence. ZBRA turns that foundation into the client's real operation.

Solution architecture

From field to decision.

01Sources

Field · sensors · systems

Field apps

GPS / RFID / QR

ERP / TMS / CRM

IoT sensors

02Ingestion

Pub/Sub

Cloud Functions

Cloud Run

Connectors

03Data Platform

BigQuery

Cloud Storage

Dataflow

Dataform

04Intelligence

Vertex AI

Predictive ETAs

Anomaly detection

Looker

05Action

Dashboards · apps · alerts

Executive dashboards

Smart apps

Real-time alerts

Mobile UIs

Google Cloud provides the technological foundation for scale, data and intelligence. ZBRA turns that foundation into operational solutions that work in the client's real context.

From extraction to reliable data

Data integrity is what holds the model up.

Before predicting, alerting or recommending, you have to ensure the data is traceable, comparable and correct. That's the quiet work that makes all the intelligence possible.

Stage 1

Extraction

APIs, field apps, queues, connectors, sensors and events.

Stage 2

Normalization

Common schemas, units, time zone, taxonomy.

Stage 3

Validation

Quality, completeness, deduplication, business rules.

Stage 4

Correlation

Events correlated by cycle: delivery ↔ route ↔ well ↔ shift.

Stage 5

Serving

Served to analytics, dashboards, ML and smart apps.

100%

operational consistency

5×

faster sync

98%

visibility accuracy

<1 min

average asset update

Intelligence layer

With the foundation ready, agentic AI steps in.

With the operational data in shape, autonomous agents work alongside ML and analytics: they monitor the operation, read the context, suggest dispatch, anticipate stoppages and run triage.

AI Agent · Operations copilot

Natural-language Q&A about the operation

AI Agent · Autonomous dispatch

Reallocates trucks and crews in real time

AI Agent · Incident triage

Diagnosis + proposed action plan

ML · Predictive maintenance

Anticipates failures in critical assets

ML · Anomaly detection

Pressures, routes, off-pattern events

Analytics · Continuous optimization

ETAs, dispatch, resource allocation

ops.zbra.dev / agents · live8 agents

Auto-resolved · 24h

+12

Agent uptime

99.8%

live

Anomalies · 24h

−38%

Ops Agent · live14 sites · 3 shifts

Which assets are at risk in the next 6h?

Found 3 assets at risk in the next 6 hours:

RIG-08Csand-out forecast 14:00P 87%

RIG-04Apressure drift +2.3σP 64%

T-117off planned routeP 52%

Reallocate trucksView plan2.1s · 4 sources

Agent activity · 24h

autonomous

DISPATCHReallocated 4 trucks · zone B → D8m

MAINTENANCEScheduled inspection · pumps RIG-04 & 1221m

TRIAGEPressure anomaly · escalated to lead47m

Results

What changes when the data shows up on time.

Every number below is a decision that only became possible because the information arrived tracked and in real time.

Asset traceability

99%: reduction in audit timefrom 3 days to 15 minutes
150k+: assets under one unified viewon a single platform
1 min: average updatenear real time

Field logistics

45%: fewer stoppages from supply shortagesproduction kept running
90%: fewer reroutesroutes planned and followed
100%: deliveries trackedin real time

Field operations

35%: less downtimea more stable operation
65%: more crew efficiencyless rework
5×: faster data syncfield → control center

“The operation started deciding on reliable, fast data, close to where events actually happen.”

A replicable capability

A capability we can rebuild for any operation.

What changes from client to client are the data sources, the process and the indicators. What doesn't change is the path: capture the information at the source, move it to the cloud and turn it into decisions.

Integrate fragmented systems

ERP, TMS, CRM, sensors, GPS, spreadsheets. The layer connects what exists without replacing it.

Capture data in the field

Offline-first apps designed for the real environment: gloves, cold and no signal.

Operate in the cloud

Ingestion, normalization, governance and analytics on a scalable platform.

Real-time visibility

Dashboards, maps, delivery status and alerts.

Applied intelligence

Predictions, anomalies and prioritization.

Operational result

Less downtime, less demurrage, more scale. Metric before model.

Applicable sectors: energy, logistics, mining, agribusiness, heavy industry, biotech, healthcare and manufacturing. Any operation where data starts in the field, lives in separate systems and has to become decisions fast.

Summary

Every decision starts with data that shows up on time.

We build the layer that connects field, cloud and action in any operation.

Talk to ZBRA