How to set up equipment fault reporting and recovery
A fault reporting routine makes sure that when a machine stops or misbehaves, the right person knows quickly, the repair is tracked, and the machine is checked before it goes back to work. This guide shows how a small plant can set it up with a few records. It matters because unrecorded breakdowns are the ones that happen again.
Why this matters
A packaging unit in Pune runs a form-fill-seal machine, two cartoning lines and a compressor. When the sealing jaw failed last month, the operator told a colleague, who told the supervisor, who told the maintenance engineer two hours later. The jaw was fixed, but no one noted the fault, the part or the time lost. When it failed again six weeks later, nobody could say it was the same fault, so the cause was never traced.
Before you start
- An asset register that gives every machine a number, a location, the make and model, and the person responsible.
- Asset numbers painted or tagged on each machine, readable from a metre away.
- A fault ticket form, with fields for asset ID, date and time, who reported, what happened and whether it is unsafe.
- A list of approved vendors for jobs the in-house team cannot do, with their contact numbers.
- A downtime log with start time, end time and the reason code for each stop.
Step by step
- The operator who notices the problem opens a fault ticket straight away, writing the asset ID, the time, and what they saw or heard. If the machine is unsafe, they stop it and mark it as unsafe on the ticket before anything else.
- The maintenance engineer looks at new tickets and uses the asset register to see the machine's history. The engineer rates the risk (safety, production stop, or minor), calls the person who should attend, and writes down who was called and when.
- The plant head decides how the repair will be done: in house, or through an approved vendor if it needs a specialist or exceeds a spending limit. The decision, the estimated cost and the expected time are recorded on a repair note.
- The repair is carried out, and the repair note is updated with what was found, what was replaced and the actual time taken. Replaced parts are noted with their source so a repeat failure can be traced.
- Before the machine goes back to work, the operator who runs it checks that it runs correctly and safely, and writes the result in the test record. Guards, switches and settings are checked as part of this, not only the output.
- The maintenance engineer closes the ticket only after the test record is signed. The downtime log gets the final start and end times, and the cause is named in plain words, such as "worn sealing jaw, heater cable loose".
- If the same asset has had the same kind of fault more than once, the maintenance engineer adds a prevention action with an owner and a date, for example a weekly check or a spare part kept in stock.
Who does what
| Role | What they do | Record they own |
|---|---|---|
| Operator | Reports the fault, stops unsafe machines, tests before restart | Fault ticket and test record |
| Maintenance engineer | Triages risk, calls the right person, closes the ticket, logs downtime | Asset register and downtime log |
| Plant head | Decides in-house or vendor repair, approves spend | Repair note |
Common mistakes
- Reporting faults by word of mouth so there is no time stamp and no history.
- Restarting a machine after repair without a test, especially after guards or electrical work.
- Writing "machine problem" as the cause, which gives no chance of spotting a pattern.
- Closing a ticket when the repair ends instead of when the machine is tested and the downtime is logged.
- Using the same vendor for every job without comparing cost and time, or without recording what they did.
How to tell it is working
- Repeat failures. For each machine, count the faults in the last 90 days that have the same cause as an earlier one. Divide by all faults on that machine. The share should fall after prevention actions begin.
- Mean repair time. Add up the hours between a ticket being opened and the machine being tested and released for all faults in the month, and divide by the number of faults. Track it by machine as well as overall.
- Unresolved faults. Count the tickets still open, and note the age of the oldest. Any ticket open beyond your own limit, for example three days, should be on the plant head's list.
Questions people ask
Who owns fault to fix?
The maintenance engineer owns the process and the history, while the plant head owns the decisions on cost and vendors. The operator owns the report and the test before restart. Every asset should also have one named responsible person.
What is the minimum record for fault to fix?
For each fault, keep the ticket with asset ID and time, the repair note, the test record and the downtime entry. With these four you can say what failed, who fixed it, how long it took and whether it was checked.
When should a case stop?
A fault case stops when the machine has passed its test, the downtime is logged and any prevention action has an owner and a date. A repair that is only temporary should stay open with the final repair planned.
Can a spreadsheet support this stream?
Yes, as long as the asset IDs are consistent and each fault is a row, never an edit of an old row. Add a sheet for the asset register and one for downtime, and a rule that the ticket number is written on the repair note.
What should the owner review?
The owner should look at the machines with the most downtime, the repeat failures, the faults open longer than your limit and the money spent on vendor repairs. A once-a-month look is enough unless something is unsafe.