Nobody knew the chamber ran warm at two in the morning until the load was refused.
An IoT monitoring dashboard puts the readings from your machines, meters and sensors on one screen. The current value, the range it is meant to sit in, and the person who is told when it goes outside. Downtime and run hours are counted from those same readings. Every number keeps its history, and the raw data exports for whoever does the analysis. The hardware is picked with you or your vendor. What we build is the screen, the rules and the record.
- Every reading proved against your own gauge before a screen is trusted
- Sensors and gateways chosen with you or your vendor, bought in your name
- Keeps recording at the site when the network drops, and fills the gap later
What goes wrong before the software.
The register gets filled in on Friday for the whole week
A supervisor is meant to write the chamber temperature twice a shift. On a bad week he writes Monday to Thursday in one sitting, from memory. The sheet looks complete and proves nothing.
The customer knew before we did
A consignment is refused at the dock, or a client rings to say the chiller you maintain has been dead since morning. You are explaining a failure you had no way of seeing, and the credit note is already written.
Nobody can say why the second shift was short
Everyone has a theory: a power dip, a die change, a long tea break, a slow operator. With no runtime record the loudest theory wins the meeting. The same hour goes missing next month.
Every machine came with its own app
The generator has a portal, the chiller has another, the tracker has a third. Three logins, three passwords, no screen that shows the site together, and none of them ever tells your own supervisor anything.
The dashboard dies with the internet
A plant line is not a home connection. When it drops the screen freezes and the readings for that hour go missing, which is reliably the hour something went wrong.
Servicing happens because three months passed
The pump is opened up on schedule while the bearing that started heating in week six fails on a Saturday. Run hours are nobody’s number. Condition is nobody’s number either.
How the work flows through it, step by step.
- 1
Deciding what is worth measuring
We walk the plant with your maintenance man and look at panels, cable routes, power points and where the signal actually reaches. Some points come back as not worth instrumenting yet. We say so at that stage rather than after the invoice.
- 2
Sensors chosen, not supplied
We do not make sensors or gateways. The probe, the meter and the gateway are picked with you or your vendor to suit the conditions on site. They are bought in your name, and fitted by an electrician who already works there.
- 3
The reading is proved before the screen exists
The new sensor runs beside your own gauge or thermometer for several shifts and the two are compared. If they disagree, the sensor moves or changes. A dashboard built on a doubtful reading is worse than none, because people stop believing the alerts.
- 4
Thresholds set by the people who get woken
Your supervisors say what counts as abnormal, how long it must last, and who hears about it on the night shift. An alarm nobody acts on turns into noise within a fortnight.
- 5
The night the link goes down
The gateway keeps reading, keeps applying its rules and sounds the hooter at the panel. Readings are held at the site meanwhile. When the line returns they upload in order and fill back into the history, so the chart has no hole where the trouble was.
- 6
Morning, and the month after
The shift review opens on run hours, stoppages with their reasons, and everything that went outside its range. Once a quarter the readings are exported for whoever does the analysis: a quality head, a consultant, or your own engineer with a spreadsheet.
What is inside.
Every module can be switched on, left out or changed to match how your team already works.
Live values on one screen
Current reading per device, grouped by line, plant or site, each with the time it last reported. One screen instead of four supplier portals.
Learn moreThresholds and duration rules
High and low limits, how long a condition has to persist, and how fast a value is moving. A door open for forty seconds is not treated like a failing compressor.
Learn moreAlerts, acknowledgement and escalation
Alerts on WhatsApp, SMS, email or the panel hooter, routed to the named person on that shift. An alarm nobody acknowledges climbs to the next name, and who acknowledged what is kept.
Learn moreDowntime and run hours
Running, idle and stopped time worked out from the readings themselves. When a stop crosses the duration you set, the operator picks a reason from a short list at the panel.
Learn moreHistory behind every reading
Trends by hour, shift, day or month, with two machines or two months set side by side. Tap on any alert. The readings around it open up.
Learn moreDevice health
Alerts when a sensor stops reporting, a battery falls low, a gateway loses power, or a value arrives that is physically impossible. A silent sensor is a fault, not a calm day.
Learn moreEdge buffering and backfill
When connectivity goes, readings are stored on the gateway at the site and uploaded in order once the link returns. The history is written back with the times they were taken.
Learn moreLocal rules and controlled outputs
A pump that stops at the set level, a lamp and hooter at the panel, an interlock with a safety condition. Manual override stays at the panel, and every automatic action is logged with its cause.
Energy, water and fuel
Consumption read from energy and water meters per machine, line or block, with generator run hours and fuel level beside them. That is usually where an argument about the bill ends.
Learn moreExport for whoever analyses it
Raw time-series readings for any period as a file, or over an API into your own tools. Retention is agreed before the first device is fitted.
Roles, sites and access
Operator, supervisor, maintenance in-charge, plant head and owner, each seeing their own machines and locations. Changing a threshold is a recorded act with a name against it.
Calibration reminders
A scheduled comparison against a reference instrument, and a reminder when a sensor is due for verification or replacement. Drift is caught by the calendar, not by a rejected load.
One system, each person sees their part.
Shift supervisor
The screen for his line, the alerts raised on his shift, and the acknowledgements he is answerable for at handover.
Maintenance in-charge
Run hours and cycle counts per machine, devices that have gone quiet, and the calibration due list. Service tasks are raised when the machine has earned them.
Plant head
Downtime by machine and by reason, shift against shift, and energy per line. He gets the argument settled with a chart instead of three opinions.
Owner or facility manager
Every site on a phone: the chambers, the pumps, the generator, and what went outside range last night.
Operator at the panel
A short list of stoppage reasons when a machine has been down past the set minutes, and the manual override that stays under his hand.
The numbers the owner asks for.
- Run hours and downtime per machine, by shift
- Stoppage reasons, counted for the month
- Every excursion outside range, with its duration and who acknowledged it
- Condition history against a batch, a consignment or a date
- Energy, water and fuel per machine, line or block
- Devices that stopped reporting, and the gap they left
- Raw readings exported for any period you name
Who this is built for.
- Manufacturers who want machine run hours and stoppage reasons instead of estimates
- Cold storage, dairy and pharma distributors who must show a condition record
- Facility and property teams running pumps, tanks, generators and meters across buildings
- Plants with equipment from several suppliers, each with its own separate app
- Sites where the internet drops often and the record still must not
- Automation firms wanting a monitoring layer above panels they already commissioned
When it is not the right choice.
- Sensors, gateways and panels are not made by us. They are selected with you or your vendor, bought in your name, and fitted on site by an electrician.
- One machine in one room that somebody walks past every hour. Look at it and write it down. That is cheaper than a sensor and a SIM card.
- No uptime or accuracy figure is promised on this page. What a device reads depends on where it is fitted and how it is calibrated, and that is settled on your site.
Adapted to your business before it goes live.
You are not handed a login and left to it. The product is set up around your records, your people and the tools you already use.
A site visit before a plan
Panels, cable routes, power points, mounting positions and network coverage are checked in person. The output is a written plan of what is practical to measure now and what is not.
Reading the machines you already own
Modbus, OPC UA, BACnet, RS-485, 4-20 mA, pulse and dry-contact inputs from PLCs, drives, energy meters and chillers, without disturbing the control system that runs the machine. Where a machine is under warranty or an AMC, the supplier’s permission comes first.
Thresholds and routing set with your people
Limits, durations and who is called on which shift are agreed with your supervisors, then tuned during a parallel run so real events fire and noise does not.
Connectivity for the site you actually have
Wi-Fi, wired, a 4G SIM or LoRaWAN. The choice is made per location. The local rules run on the gateway whichever one you pick, so a dropped line never becomes a plant problem.
Handover and support
Source code, device inventory, wiring and configuration documents and an operator runbook are handed over, with cloud accounts and SIMs in your name. Support afterwards covers device health, renewals, calibration and new measurement points.
Works alongside
- Modbus and OPC UA devices
- PLCs, drives and chiller controllers
- Energy and water meters
- SMS and email
- ERP and maintenance software
- Panel hooters and indication lamps
Questions owners ask before a demo.
Can the IoT monitoring dashboard keep working when the plant internet drops?
Yes, because the rules run on the gateway at the site and not in the cloud. Readings are buffered locally, the hooter still sounds, and an interlock still fires. When the link returns the stored readings upload in order and fill back into the history. We prove it during commissioning by pulling the cable out on purpose.
Do you supply the sensors and gateways?
No. We do not manufacture hardware, and we would rather say that plainly than sell you a kit. Devices are selected with you or your vendor for the conditions on your site, and bought in your name. Where a supplier portal you already own does most of this, a free technology audit will tell you that instead of quoting for a deployment.
Can it read our existing PLCs, drives and energy meters?
Usually yes. Modbus, OPC UA, BACnet, RS-485, 4-20 mA and pulse inputs cover most of what sits in an Indian plant room. We read alongside the control system rather than into it, so the machine keeps running exactly as it does today.
How accurate will the readings be?
No figure is claimed here, and you should distrust anyone who does claim one on a website. Accuracy depends on the device, where it is mounted and when it was last calibrated. What we do is run it beside your own gauge for several shifts, show you both sets of numbers, and set a calibration reminder afterwards.
Can we export the sensor data for our own analysis?
Yes. Raw readings for any period download as a file, and an API serves them into your own tools. Quality heads, consultants and insurance assessors all ask for this at some point, so retention is agreed before the first device goes up.
Can supervisors see machine remote monitoring screens on a phone?
Yes. It runs in a browser on phones and tablets, with each role seeing its own machines and sites. Alerts arrive on WhatsApp or SMS, and acknowledging one from the phone is enough to stop the escalation.
Can we bring in the readings we already have?
Exports from an older logger or a spreadsheet load as history with their timestamps. Handwritten registers are a judgement call. Typing three years of a logbook rarely pays for itself, and we will tell you so rather than bill you for it.
Does this connect to Tally or our ERP?
Run hours, downtime and consumption figures can be pushed into your ERP or maintenance software so production planning reads them. Tally keeps books, not machine logs, so the usual link there is a consumption or production figure your accountant asks for, agreed with him during set-up.
Who owns the devices, the SIMs and the data?
You do. Devices are bought in your name, cloud accounts stay in your accounts, and the readings are yours. Source code and device documentation are handed over at the end of the work.
What does an IoT monitoring dashboard cost?
No price is published. It moves with how many points are measured, how often each is read, how long readings are kept, and how much sits at the edge. Hardware is bought separately in your name. A demo on one machine usually comes before any proposal.
Businesses that use this.
Related products.
See IoT Monitoring Dashboard with your own data.
Tell us how you run things today. We’ll show you the parts that matter to you and say plainly what would need changing.
- No obligation
- We reply the same working day
- Your details stay private