01
Downtime reasons on one line
Stops captured with a reason instead of a blank.
Manufacturing / Smart Factory Pilot
Smart factory pilot: scope one line, prove it, then decide
Many smart factory demos look good in a meeting room and stall on the plant floor, because nobody agreed what success meant. Agnotic Technologies helps plants scope a pilot on one line or cell, measure a baseline first, and reach a clear decision to scale, extend or stop.
One line
Pilot scope
Baseline first
KPIs per ISO 22400
Go or no-go
Decision at the end
Trusted by global innovators
Industry Overview
What is a smart factory pilot?
A smart factory pilot is a time-boxed trial of connected equipment data, analytics or automation on one production line or cell, judged against a baseline the plant measured beforehand. Agnotic Technologies scopes and builds pilots so a smart factory demo becomes a decision: scale it, extend the trial or stop, on criteria the plant set in advance.
Industry Challenges
Why do smart factory pilots stall?
Most stalled pilots were not technical failures. They were scoping failures.
The pilot line was chosen because it was convenient, not because it has the problem the plant most wants solved.
No baseline was measured, so after the pilot nobody can say whether downtime or scrap actually changed.
Success was defined as the dashboard working, not as an operational result the plant manager cares about.
There were no stop criteria, so the pilot carries on as a permanent trial.
The pilot was built in a way that cannot scale: hand-wired connections, a laptop under a desk, data nobody else can reach.
Operators and maintenance were not asked what they need, so the screens go unused after the demo.
Decision
How do you decide whether to scale, extend or stop?
At the final review, the result against the baseline and the state of the build decide the outcome. This is the decision table we use with the plant.
| Decision | When the evidence shows | What happens next |
|---|---|---|
| Scale | The success criteria were met against the baseline and the scale-out readiness list is complete | Plan the rollout line by line, reusing the pilot's data model and connectors |
| Extend | Results look promising, but the measurement window missed normal running or one data source was unreliable | Extend the window once, with a fixed end date and the same criteria |
| Rework | The result is real, but the build cannot scale: manual steps, unsupported hardware or no owner | Fix the build on the pilot line before any second line |
| Stop | The criteria were not met, or a stop criterion was triggered | Record what was learned, remove the pilot hardware and keep the baseline data |
Our Approach
How should a plant approach a first pilot?
Three rules keep a pilot small enough to finish and serious enough to decide on.
One problem, one line
Pick the use case first, then the line where it hurts most and can be measured. A second use case waits for the second pilot.
Measure before you change
Record the baseline with the same KPI definitions the pilot will use, over a window the plant agrees represents normal running.
Build the pilot the way you would build line two
Use the connectors, data model and network placement you would use at scale, so a positive result does not need a rebuild.
Capabilities
What do we build in a smart factory pilot?
Only what the use case needs. Platform depth lives on the industrial IoT and MES pages; a pilot uses a thin slice of it.
Edge connection for the pilot line
One edge gateway reading the agreed tags or historian points, placed in the network zone the plant approved.
Baseline and pilot KPIs
The ISO 22400-based KPIs from the checklist, calculated the same way before and during the pilot.
Line screens
One view for operators and one for the plant manager, each showing the KPI against its baseline.
Use case logic
The analytics, alert or model under test, such as a downtime reason prompt or an early warning on one machine.
Decision report
A final comparison of pilot and baseline, with the evidence for the scale, extend, rework or stop decision.
Scale-out notes
What changes for the next line: tags, network, hardware, training and support.
Engineering Stack
Which standards keep a pilot comparable?
Shared definitions matter more than tools when a result has to be trusted.
KPIs and models
ISO 22400-2 KPI definitionsIEC 62264 (ISA-95) equipment hierarchyWork unit naming agreed with the plant
Connectivity
OPC UAModbus TCPHistorian interfacesMES records where they exist
Edge and platform
Industrial edge gateway on DIN railTime-series storageAWS, Microsoft Azure, Google Cloud or on premises
Security
Placement agreed with the plant's OT security ownerRead-only data access by defaultNIST SP 800-82 guidance for OT
Measured Impact
What should the pilot report at the final review?
Four measures the plant reads before deciding.
Baseline delta
Change against the baseline
The use case KPI, pilot window versus baseline window.
Data availability
Usable data in the window
Gaps and bad readings, so a result is not judged on partial data.
Adoption
Use of the pilot screens
Whether operators and maintenance actually used what was built.
Scale readiness
Readiness items closed
Open items that would block a second line.
Use Cases
Which use cases suit a first pilot?
Problems that are common, measurable on one line and visible to the plant manager.
02
Early warning on one machine
A signal before a known failure mode on a critical asset.
03
Scrap at one station
Scrap counts and causes tied to the time they happened.
04
Changeover time
Changeovers measured from last good part to first good part.
05
Energy per unit
Line energy use tied to output.
06
Manual log replacement
Paper line logs moved to structured data.
Integrations
Which plant systems does a pilot touch?
Usually read-only, and only the ones the use case needs.
Machine signals
PLCs
Time-series data
Historian
Orders and counts
MES
Maintenance history
CMMS
Products and orders
ERP
Scrap and defects
Quality system
What we check
What goes on a smart factory pilot scoping checklist?
The checklist we complete with the plant before any hardware is ordered. Each line ends in a written decision, not an open question.
Use case and line
- One use case, stated as the operational problem: unplanned downtime, scrap, changeover time or energy per unit
- The line or cell where that problem is visible and measurable today
- Assets in scope listed by ID, and assets explicitly out of scope
Data sources
- Each source named: PLC tags, historian points, MES records or manual logs
- Protocol per source, such as OPC UA, Modbus or a historian interface, confirmed with controls engineering
- Who owns access to each source and who approves the connection
Baseline
- KPIs taken from ISO 22400 definitions, so the plant and the build calculate them the same way
- Measurement window agreed, covering normal shifts and product mix
- Baseline values recorded and signed off before the pilot changes anything
Network zone
- Zone for the edge device agreed with the plant's OT security owner
- Read-only access unless a write path is part of the use case and approved
Success and stop criteria
- Success criteria the plant sets, written as a change against the baseline
- Stop criteria: the result, cost or disruption that ends the pilot early
- Who makes the final decision, and when
Decision and scale-out
- Decision options written down: scale, extend, rework or stop
- Scale-out readiness list reviewed before the final review
Compliance-First Development Services Backed by Global Standards
We build secure, scalable products designed for privacy, interoperability, and regulatory readiness from day one across every sector we serve.
Information Security Management
Adhering to the international gold standard for managing information security risks.
Quality Management Systems
Ensuring consistent quality in software delivery and operational process optimization.
Our Edge
Why Global Leaders Choose Us
We combine deep technical expertise with industry-specific knowledge to deliver solutions that aren't just functional, but transformational.
Enterprise-Grade Security
We implement rigorous security protocols and compliance standards (HIPAA, GDPR, SOC2) across all industrial solutions to protect sensitive data.
High-Performance Scaling
Our architectures are built to handle massive data loads and user bases, keeping performance steady whether you're serving ten or ten million.
Accelerated Time-to-Market
Using our suite of internal tools and proven frameworks, we cut setup work so your first release reaches users sooner.
Embedded AI Integration
Beyond simple wrappers, we build deep-learning integrations and predictive analytics directly into the core of your industry-specific workflows.
Engagement Model
How does a smart factory pilot engagement run?
Smart factory demo services that end in a decision. It depends on scope; we agree the first release and the measurement window on the scoping call.
01
Scoping workshop
We complete the pilot scoping checklist with operations, maintenance, controls and OT security.
02
Baseline
The agreed KPIs are measured on the pilot line before anything changes.
03
Build
The edge connection, KPI calculations, line screens and use case logic, built the way line two would be.
04
Pilot run
The pilot runs through the measurement window the plant agreed.
05
Decision review
Results against the baseline and the readiness list lead to a scale, extend, rework or stop decision.
Facts and sources
Standards behind a comparable pilot
The KPI, model and security references the checklist relies on.
ISO 22400 sets an industry-neutral framework for defining, composing, exchanging and using key performance indicators for manufacturing operations management in batch, continuous and discrete industries.
Source: ISO 22400-1:2014 (ISO)ISO 22400-2:2014 presents a selected set of manufacturing KPIs by their formula and elements, time behaviour, unit and the user group and production methodology they apply to, and relates them to work units as specified in IEC 62264.
Source: ISO 22400-2:2014 (ISO)IEC 62264-1:2013 describes the manufacturing operations management domain (Level 3) and the interface content between it and the enterprise domain (Level 4), with the goal of reducing the risk, cost and errors of implementing those interfaces.
Source: IEC 62264-1:2013 (ISO)The OPC Foundation describes OPC UA, released in 2008, as a platform-independent service-oriented architecture with encryption, authentication and auditing, and an extensible information modeling framework.
Source: OPC Foundation, OPC Unified ArchitectureNIST SP 800-82 Rev. 3, published in September 2023, gives guidance on securing operational technology, including industrial control systems, while addressing its performance, reliability and safety requirements.
Source: NIST SP 800-82 Rev. 3 (NIST CSRC)
Voice of success
We don't just build products; we forge lasting partnerships. See how we've helped industry leaders transform their vision into technical reality.
“I can clearly see how Agnotic has a unique way of handling end-to-end development. They are always active on quick chat and provide support quickly.”
Aaron Phelan
Founder, Benchmark
“Agnotic combines deep technical expertise with strong domain knowledge. They understand the business context, anticipate challenges, and make collaboration smooth and effective.”
John Pasmore
Founder, Latimer
“Agnotic is the best technical team we evaluated. Their engineering excellence made our work dramatically easier and allowed us to stay focused on what matters most for maternal care outcomes. They took full ownership of the technical execution, and we are always happy to continue working together.”
Kim Smith
Founder, My Lauren
Related pages
Related manufacturing pages
This page covers scoping and judging a pilot. These pages cover the platforms, connectors and calculations it uses.
Industrial IoT data platform
OT connectors, asset data model and the platform a successful pilot grows into.
Manufacturing execution system
Custom MES for production tracking, genealogy and quality gates.
OEE software
How availability, performance and quality are calculated and tracked continuously.
SCADA integration services
Protocol connectivity and the OT and IT boundary, including network segmentation.
MES integration services
Connecting MES with ERP and machine data.
AI proof of concept
Feasibility work for pilots whose use case depends on a model.
More in Manufacturing
More Manufacturing Solutions
Explore other production-grade engineering platforms we deliver across manufacturing.
Frequently Asked Questions
A demo shows that the technology works, often on sample data. A pilot runs on a real line against a baseline the plant measured, with success and stop criteria agreed in advance, and ends in a decision to scale, extend or stop.
Start from the problem, not the line. Choose the line or cell where the problem is visible, the data can be reached, the team is willing and a result would matter to the plant. Avoid the most unusual line, because results there rarely transfer.
ISO 22400 definitions mean the plant, the pilot and any later supplier calculate KPIs the same way, with the same elements and time behaviour. Without a shared definition, a pilot result can be argued away.
The plant does, usually the plant manager with operations, maintenance and quality leads. We help write them as measurable changes against the baseline, but we do not set the bar our own work is judged against.
Usually yes. Most pilots only read data, through an edge gateway placed in the zone your OT security owner approves. Writing back to controls is added only when the use case needs it and the plant approves it.
The number of assets and data sources, the state of existing connectivity, whether new sensors are needed, the use case logic and the length of the measurement window. We explain these drivers on the scoping call instead of quoting a fixed price.
It depends on scope; we agree a fixed first release and the measurement window on the scoping call. The window has to cover normal shifts and product mix, so the plant's production pattern sets it, not our build.
We review the scale-out readiness list with the plant: data model, connectors, network placement, support and training. If those are in place, rollout continues line by line on the same build, usually onto an industrial IoT data platform.
Industries