Downtime tracking
Machine downtime tracking that does not depend on memory
LiveMach machine downtime tracking is an industrial IoT tool for Indian manufacturers. Its own sensors, fitted outside the machine, record every stop with its start and end time. The operator picks the reason on the dashboard. You see downtime by machine, reason and shift. It is for plant heads, production managers and factory owners.
Downtime tracking is only useful when the timing comes from the machine and the reason comes from the person standing next to it. LiveMach records the first automatically and makes the second a two tap job on the dashboard.
What is the difference between planned and unplanned downtime?
Planned downtime is time you decided not to run: lunch, tea, a scheduled maintenance slot, a shift that is not staffed. Unplanned downtime is time you meant to run and could not. They are two different problems, and mixing them hides the one you can actually fix.
Most plants report one number called "downtime" and it includes both. Then a 45 minute changeover, which is planned work that can be shortened, sits in the same bucket as a 25 minute coolant pump failure, which is a maintenance problem. The two need completely different responses, and averaging them means neither gets attacked.
| Type | Examples | The question it raises |
|---|---|---|
| Planned | Lunch and tea, planned preventive maintenance, an unstaffed third shift, a public holiday. | Is the plan right? Should this slot exist at all? |
| Semi planned | Changeover, die or mould change, tool change, first-off approval waiting for QC. | It has to happen. How much shorter can we make it? |
| Unplanned | Breakdown, no material, no operator, power cut, quality hold, no work order. | Why did this happen, and what stops it happening again? |
We keep changeover separate on purpose. In a job shop it is usually the single biggest block of lost time, and calling it a breakdown makes it invisible to the people who could shorten it.
Which downtime reason codes fit an Indian plant?
A reason code list works when an operator can pick the right one in under three seconds. Start with eight or nine codes, not thirty.
This is a realistic starting list. Change the words to match what your floor already says.
No material
Raw material, component or consumable not at the machine.
No operator
Nobody at the machine. Absent, on another machine, or on break beyond the planned slot.
Tool change
Insert, cutter, blade or screen change during a run.
Die or mould change
Job change on a press, moulding machine or printing line.
Breakdown
Machine fault that needs a fitter or an electrician.
Power cut or DG changeover
Supply failure and the gap while the DG set takes load.
Quality hold
Machine stopped for inspection, first-off approval or a rejection investigation.
No work order
Machine free and available, nothing scheduled on it.
Lunch and tea
Planned break. Kept separate so it never pollutes the unplanned numbers.
Two rules that keep the list honest
- Never let "Other" grow. If it crosses one in ten stoppages, you are missing a code your floor actually needs. Add it.
- Do not add a code your operator cannot tell apart from an existing one in three seconds. A precise list nobody can use is worse than a rough list everybody can.
Which reasons cause most of the lost time?
In almost every plant we have measured, a small handful of reasons account for the majority of downtime. Finding which ones is the entire value of tracking.
This is the ordinary Pareto pattern. It sounds obvious written down, and it is still the thing that surprises people most, because the reason everybody in the review meeting blames is usually not in the top two.
Illustrative pattern, not client data
One week of downtime on a 12 machine shop
Reasons ranked by minutes lost, with the running total drawn across them. Two reasons out of eight account for 60 percent of the loss.
| # | Reason | Minutes lost | Running total |
|---|---|---|---|
| 1 | Changeover | 620 min | 36% |
| 2 | No material | 430 min | 60% |
| 3 | Breakdown | 250 min | 75% |
| 4 | No operator | 160 min | 84% |
| 5 | Power cut or DG | 110 min | 90% |
| 6 | Quality hold | 70 min | 94% |
| 7 | Tool change | 60 min | 98% |
| 8 | No work order | 40 min | 100% |
| Total lost | 1740 min |
Once you can see the ranking, the argument changes. Instead of "the machines keep breaking down", the conversation becomes "changeover is costing us ten hours a week on these four machines, what will it take to halve it".
Can I track downtime in Excel?
A spreadsheet is a sensible place to start, and plenty of plants run one well for a year. It teaches you what to measure before you spend anything.
So we are not going to tell you to throw it away. We will tell you exactly where it stops working, and you can decide whether you have reached that point.
Where the spreadsheet stops working
It runs on recall
The entry is made at the end of the shift by someone whose job was to run the machine, not to watch the clock. Short stops are forgotten and long ones get rounded.
No timestamps, so no patterns
You know 90 minutes were lost. You do not know that 60 of them were between 2 and 4 pm, every single day, which is when the material trolley does not arrive.
Reason discipline slips
In month one the codes are clean. By month three "Breakdown" and "Other" are half the entries, and the Pareto is measuring nothing.
One person holds it together
It works because a supervisor or a planner cares about it. When they take leave the sheet has a gap. When they leave the company it usually dies within two months.
The night shift is invisible
This is the one that matters most. The night shift is where the losses are, and it is the shift with the least supervision and the softest reporting.
If your sheet is clean and your reviews are working, keep going. When you notice the meeting has turned into an argument about whether the data is right, the sheet has done its job.
How does LiveMach record a stoppage?
Our sensors read the machine directly, so the moment it stops and the moment it starts again are recorded on their own, to the minute, on every shift including nights.
1
The machine stops
Our device sees the state change from running to stopped. The clock starts. Nobody has to notice or write anything.
2
The reason gets attached
The operator picks a reason on the dashboard, in English or Hindi. Because the stoppage is already recorded and waiting, all that is needed is the reason itself.
3
It survives a network drop
If your internet is down, the device keeps logging to its own memory and syncs everything when the connection returns. A drop does not cost you a shift of data.
4
The report writes itself
Shift, day and month reports are ready when you open the dashboard. Downtime by machine, by reason, by shift and by operator, with the timestamps intact.
The split matters. The machine reports the timing, which is the part people get wrong. Your operator supplies the reason, which is the part only a person on the floor knows.
How do you find out where your hours are going?
Start with a pilot on one machine or one line. Two weeks of real timestamps usually settles an argument that has been running for years.
Common questions
- Will production stop while you install it?
- No. Our sensors sit outside the machine and nothing inside is cut or rewired, so fitting is usually done during a normal changeover or a shift gap. On most machines the device is on and reporting within an hour.
- Will it work on my 1998 machine with no digital output?
- Yes. This is exactly the case we designed for. We read the machine from outside using sensors on indicator lamps, contactors, the spindle drive current or a proximity sensor on a moving part. The machine does not need a controller, a network port or any digital output.
- What happens if the internet goes down?
- The device keeps logging to its own memory and syncs automatically when the connection returns. You do not lose the stoppages that happened during the outage, and the timestamps stay correct.
- What if your device fails?
- Your machine keeps running. The device is electrically isolated and sits alongside the machine rather than in its control path, so it can never stop production. You lose visibility until it is replaced, not output.
- Do operators have to enter every reason?
- They enter the reason, not the timing. The stoppage, its start, its end and its duration are already recorded by the machine. The operator only picks a reason code from a short list, which takes a couple of seconds on the dashboard.
- What does downtime tracking cost?
- It depends on how many machines you fit, which sensors each machine needs and whether you want add-ons like power monitoring. Every deployment is quoted after we understand your machines, so tell us your machine count and types and we will give you a real number.
- Who maintains the system once it is installed?
- We do. The dashboard is hosted and updated by us, and firmware on the devices updates over the air. Your team uses it, picks reason codes and reads the reports. Nobody on your side has to look after servers.
- Who is machine downtime tracking for?
- It is for plant heads, production managers, maintenance heads and owners of small and mid size factories in India who want to know how long machines stop and why. It is not an MES or an ERP. It is a monitoring tool.
Related reading
How to track machine downtime
A step by step guide, from a paper register to automatic tracking.
OEE software
Downtime is the Availability factor in OEE. This page works the whole calculation through on a real shift.
CNC machine monitoring
The same four states, read off a CNC without touching the controller.
Stack light monitoring
How the lamps on a signal tower tell us if a machine is running or stopped.
The LiveMach sensor device
What the device reads, how it is isolated and what happens when the internet drops.
How LiveMach pricing works
What decides the cost and how a pilot on one machine starts.