01
One-cell pilot
A handful of machines connected, states agreed and reasons captured.
Manufacturing / Machine Monitoring
Machine monitoring software for live status and downtime reasons
Many plants still run on end-of-shift guesses about why machines stopped. Agnotic Technologies builds machine monitoring software that connects old and new machines, turns signals into clear states, asks the operator for a reason at the machine and calculates OEE from what actually happened.
States
Running, idle, setup, down
Reasons
Captured at the machine
OEE
From states, not estimates
Trusted by global innovators
Industry Overview
What is machine monitoring software?
Machine monitoring software collects signals from machines on the shop floor and turns them into live status, downtime records and OEE. Agnotic Technologies builds it for plants that want this before, or without, a full MES: connections to machines of any age, state rules, operator reason codes, alerts and shift reports.
Industry Challenges
Why are downtime numbers usually wrong?
Not because people are careless. Because the data is collected the wrong way.
Downtime is written on a sheet at shift end, so short stops disappear and long ones are rounded.
Newer CNC machines report plenty of data while older machines report nothing, so the dashboard covers half the floor.
A machine that is powered on but waiting for material counts as running.
The reason code list has dozens of options, so operators pick the first one.
OEE is calculated in spreadsheets, with a different definition in each plant.
Alerts go to everyone, so nobody acts on them.
Decision table
How do we connect each type of machine?
Most plants have all three. We survey each machine and record which row it falls in before we estimate the work.
| Machine type | Connection method | Signals available | State accuracy |
|---|---|---|---|
| CNC machine with an MTConnect agent or OPC UA for Machine Tools | Read the agent or OPC UA server over the plant network | Execution state, program, part count, spindle and feed values, alarms | High: states come from the controller itself |
| PLC-controlled line or cell | Read PLC tags through an OPC UA server or a protocol gateway | Run, fault and mode bits, counters, cycle complete signals | High when the tags exist; depends on what the PLC program exposes |
| Legacy machine with no usable controller | Retrofit current sensors or a stack light tap on an edge device | Power draw, stack light colour, and a part sensor where fitted | Moderate: states are inferred from load, so thresholds are tuned per machine |
Our Approach
When should a plant start with machine monitoring instead of an MES?
When the main question is why machines stop, not how orders flow through the plant.
Start small
Monitoring answers availability and downtime questions with only machine connections and a reason screen. No routings, bills of materials or order integration are needed.
Keep it MES-ready
Machines, states and reasons are modelled so an MES can use them later, instead of becoming another island of data.
Grow when orders matter
When you need work orders, genealogy and material tracking, that is the point to move to an MES, and the monitoring data carries over.
Capabilities
What do we build in machine monitoring software?
We build the monitoring layer. Failure prediction is covered on our predictive maintenance page, and plant-wide analytics on our IIoT data platform page.
Machine connectors
MTConnect, OPC UA, PLC tags and retrofit sensors, read-only on the controller side.
State engine
Configurable rules that turn signals into running, idle, setup, down and planned stop.
Reason code capture
A short tablet screen at the machine that asks for a reason while the machine is stopped.
OEE calculation
Availability, performance and quality from states, counts and rejects, with one documented definition.
Alerts
Notifications to the right person by machine, state and duration, not to everyone.
Shift reports
Downtime by reason, machine and shift, ready at the end of each shift.
Live floor view
A plant layout showing each machine's current state and how long it has been in it.
Data export
States and reasons sent to your warehouse or IIoT platform for wider analysis.
Engineering Stack
Standards, edge and data tools for machine monitoring
Chosen per machine type and per plant network.
Machine standards
MTConnectOPC UA and OPC UA for Machine ToolsModbus TCP and EtherNet/IP through gateways
Edge
Edge devices for retrofit sensorsStore and forward during network lossTime synchronisation
Data
Time-series databasePostgreSQL for states and reasonsShift calendar service
Apps and security
Web floor viewTablet reason screenRole-based accessSegmentation from the control network
State rules
How do signals become machine states?
Rules are set per machine and reviewed with the people who run it. These are typical starting rules.
| State | Example rule | Who confirms it |
|---|---|---|
| Running | Controller execution is active, or spindle load stays above the idle threshold for longer than a set time | Production lead for the cell |
| Idle | Powered and ready, but no cycle running and no reason entered yet | The operator, by entering a reason |
| Setup | The operator selects setup at the machine, or a program change is detected with no parts counted | The operator at the machine |
| Down | A fault or alarm is active, or the machine has stopped longer than the idle limit with a downtime reason | The operator enters the reason; maintenance closes it |
| Planned stop | Inside a break, meeting or planned maintenance window from the shift calendar | The shift calendar owner |
Measured Impact
What does machine monitoring report?
Measures that come straight from states and reasons.
Availability
Run time against planned time
Per machine, cell and shift.
Top reasons
Downtime by reason code
Where the stopped hours actually go.
Short stops
Stops under the idle limit
Small losses that never reach a paper log.
Unexplained
Downtime without a reason
Stops still waiting for an operator reason.
Use Cases
Where do machine monitoring projects start?
First releases we scope, each small enough to prove on one part of the floor.
02
CNC shop connection
Machines with MTConnect or OPC UA connected across a machining area.
03
Legacy machine retrofit
Current sensors and stack light taps on machines with no usable controller.
04
Reason code redesign
A short, machine-specific reason list that operators actually use.
05
Multi-plant OEE standard
One documented OEE definition applied the same way in every plant.
06
Machine monitoring for startups
Equipment makers and startups that want monitoring built into the machines they sell.
Integrations
Which systems does machine monitoring connect to?
Machines on one side; planning, maintenance and analytics on the other.
Machine data
CNC controllers
Line data
PLCs
Retrofit sensors
Edge gateways
Planned orders and cycle times
ERP
Work orders
Maintenance system
Production
MES
Analytics
IIoT data platform
Alerts
Messaging
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 machine monitoring engagement run?
We start with a machine survey and one cell. We agree a fixed first release on the scoping call.
01
Scoping call
Machines, controllers, network rules and the downtime questions you need answered.
02
Machine survey
Each machine placed in the connection table, with the signals it offers.
03
Fixed-scope first release
One cell connected, states and reasons agreed, OEE reported.
04
Roll out
More machines and areas on the same model, legacy retrofits included.
05
Support
New machines, rule changes and reason list reviews under agreed support.
Facts and sources
Which standards shape machine monitoring?
Short facts we design around, each linked to its primary source.
The MTConnect REST API is stateless and read-only, and it does not produce any side effects on the agent or the equipment it reports on.
Source: MTConnect Standard Part 1, Fundamentals, version 2.5.0MTConnect's Execution data item reports a component's operating state with values such as ACTIVE, READY (idle and ready to execute), INTERRUPTED, FEED_HOLD, PROGRAM_STOPPED and STOPPED.
Source: MTConnect Standard Part 3, Observation Information Model, version 2.5.0OPC UA for Machine Tools, Part 1, specifies an information model for machine monitoring and job management, giving a common interface across machine tools of different technologies, manufacturers and model series for MES, SCADA, ERP and analytics systems.
Source: OPC Foundation and VDW, OPC 40501-1, OPC UA for Machine Tools Part 1ISO 22400-2:2014 specifies a selected number of manufacturing operations KPIs, each presented with its formula and elements, time behaviour and unit, and relates them to work units as specified in IEC 62264.
Source: ISO 22400-2:2014, Key performance indicators for manufacturing operations management, Part 2
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 software services
This page covers machine status and downtime. These pages cover the systems that build on it.
Manufacturing execution system
Work orders, genealogy and quality gates, when monitoring is no longer enough.
Industrial IoT data platform
Plant-wide data and analytics beyond machine states.
Predictive maintenance dashboard
Failure prediction from condition and machine data.
MES integration services
Connecting machines and MES to ERP and other systems.
Manufacturing software development
Our wider manufacturing practice.
More in Manufacturing
More Manufacturing Solutions
Explore other production-grade engineering platforms we deliver across manufacturing.
Cross-Industry Reach
The Same Engineering Patterns, Across Other Industries
The technical patterns behind this platform translate naturally into adjacent verticals.
Frequently Asked Questions
Machine monitoring tells you what each machine is doing and why it stopped. An MES also runs production: work orders, routings, material and genealogy. Many plants start with monitoring and add an MES later on the same data.
Yes, with retrofit current sensors or a stack light tap on an edge device. States are inferred from load, so we tune thresholds per machine and confirm them with the people who run it.
The number and mix of machines, how many need retrofit sensors, network work on the shop floor, the reason screens and the systems to connect. The machine survey is what turns these drivers into an estimate.
We connect read-only. MTConnect's interface is read-only by design, OPC UA connections use read access only, and retrofit sensors sit outside the controller entirely.
From states, counts and rejects: availability from run time against planned time, performance from actual against ideal cycle time, and quality from good parts against total. We document one definition, using ISO 22400-2 as the reference where you adopt it, and apply it in every plant.
No. It records states and downtime reasons. Predicting failures needs condition data and models, which is covered on our predictive maintenance dashboard page.
Industries