Key points
- Breakdown analysis looks across many failures to find where time and money go, while root cause analysis digs into one failure to find why it happens.
- The analysis is only as good as the log: equipment from the register, stop and restart times, a breakdown code, parts and labour, filled in the same way every time.
- Rank equipment by count, downtime and cost separately, because a machine that stops often and a machine that stops for long are different problems.
- Every item at the top of the list needs one action, one owner and a date, and next period's numbers show whether it worked.
Breakdown analysis vs root cause analysis
A single breakdown tells you something broke. Fifty breakdowns, sorted properly, tell you where your time and money are going. Breakdown analysis is that second kind of work: it looks across all the failures in a period, usually a month, a quarter or six months, and asks which equipment, which failure types and which locations or shifts account for most of the lost hours.
Root cause analysis works the other way. It takes one failure, or one repeating failure, and digs down until it finds why it happens. The two fit together: breakdown analysis picks the target, root cause analysis takes it apart. Running a full root cause analysis on every breakdown wastes effort, and skipping breakdown analysis means you choose targets by whoever complains loudest.
What to record on every breakdown
You do not need many fields, but you need the same fields filled in every time. Missing or free text fields are what make most breakdown logs impossible to analyse later.
| Field | Why it matters |
|---|---|
| Equipment, chosen from the asset register | Lets you group by machine. Free text such as "the big conveyor" cannot be counted reliably. |
| Time the equipment stopped and time it restarted | Gives downtime, usually the number that matters most. |
| Breakdown code (failure type) | Lets you group by what failed: belt, bearing, sensor, electrical and so on. |
| Short description and what was done | Gives context when the code alone is not enough. |
| Parts used and labour hours | Gives the repair cost of each breakdown. |
| Whether production stopped | Separates failures that halted output from those the team worked around. |
How to run a breakdown analysis, step by step
- Pick a period long enough to show patterns. A quarter is a reasonable start for most teams; a single month often has too few breakdowns per machine to say anything.
- Clean the data first. Fill in missing restart times, merge duplicate records and map any free text equipment names to the asset register.
- Group by equipment and add up three things: number of breakdowns, total downtime hours and total repair cost.
- Sort each of the three from highest to lowest. Look at the top of each list, not at the averages.
- Pick the top two or three by downtime, or by cost for equipment whose stops do not halt anything, such as standby units.
- For those few, group again by breakdown code to see which failure types repeat.
- Decide one action for each: root cause analysis, a new or changed preventive maintenance task, an operator check, a spare part to stock, or replacement.
- Repeat next period with the same lists and compare.
Count and downtime often point to different machines. A machine that stops many times for a few minutes is irritating; a machine that stops twice a quarter for a full day may be the bigger loss. Look at both lists before deciding where to spend effort.
Worked example: one quarter on a small production line
This is an illustrative example with made up figures, to show how the sorting works in practice.
| Equipment | Breakdowns | Downtime (hours) |
|---|---|---|
| Conveyor 3 | 14 | 21 |
| Packing machine 2 | 4 | 38 |
| Air compressor 1 | 6 | 9 |
| Forklift 2 | 9 | 6 |
| Pump 4 | 2 | 3 |
Total downtime is 77 hours. By count, Conveyor 3 looks like the problem. By downtime, the packing machine is, with 38 hours from only four stops. Together these two account for 59 of the 77 hours, so they become the focus for next quarter, and the other three get a quick review only.
Now drill into each. Grouping the conveyor's 14 breakdowns by code might show 8 belt tracking faults, 4 sensor faults and 2 motor trips. Belt tracking is the repeat failure, so it becomes the root cause analysis target, and a weekly belt tracking check may go into the preventive plan. The packing machine's four stops might all be different, but each one long. There the useful question is why repairs take so long: waiting for parts, missing skills or no written procedure. That points to spare parts stock or a repair guide rather than to one failure mode.
Keep breakdown codes short and usable
Breakdown codes are what make grouping by failure type possible. With too few, everything ends up as "mechanical". With too many, technicians pick the first one in the list.
- Start with ten to fifteen codes that match how your equipment actually fails, for example belt, bearing, seal or leak, electrical, sensor, hydraulic, pneumatic, operator error and external cause.
- Always include "other" and read what lands there each month. If the same thing keeps appearing, give it its own code.
- Let the code say what failed and the description say why. Mixing the two makes the list grow without limit.
- Make the code required when the breakdown is closed, not when it is opened. The technician only knows the failure after looking at it.
- Do not rename or merge codes in the middle of a period, or the comparison with the last period breaks.
Turning the analysis into action
An analysis that ends as a chart has not done its job. For each item at the top of the list, write one action, one owner and a date, and raise it as a work order so it is tracked like any other job.
| What you see | Likely action |
|---|---|
| The same failure repeating on one machine | Root cause analysis, then a preventive task or a design change |
| The same failure across many similar machines | Change the preventive plan for that whole equipment type |
| Few stops, but very long repairs | Stock critical spares, write a repair procedure, check skills |
| Many short stops linked to how the machine is run | Operator training or a daily checklist |
| Repair cost climbing on an old machine | Compare running cost with the cost of replacement |
Next period, check breakdown count, downtime and MTBF for the same equipment. If the numbers have not moved, the action did not reach the real cause, and the item goes back on the list.
If breakdowns are logged in a CMMS such as Solmex against the asset register, with codes and stop times, the grouping in steps 3 to 6 becomes a report rather than a spreadsheet job, and you can rerun it the same way every period.