Why rank equipment at all
A plant may have thousands of assets and a limited number of technicians, hours and dollars. Treating every machine identically means spending too much on trivial items and too little on the ones that can stop production or hurt someone. Criticality ranking is a structured way to decide where effort should go. It sorts equipment by how serious the consequences of a failure would be, so that inspections, spares, monitoring and repair priority follow that order.
The key word is consequences. Criticality is not about whether a machine is old, noisy or unpopular with the crew. A very reliable pump can still be critical if its failure would shut the plant, and a troublesome small fan can be low in criticality if nothing depends on it.
Criticality is not the same as condition
Two separate questions are often mixed together.
- How bad would a failure be? This is criticality.
- How likely is a failure? This depends on the machine’s condition, age, duty and history.
Ranking uses the first question. The second is considered afterward, when deciding what to do about the assets near the top. A critical machine in good condition may only need a regular inspection. A less critical machine that fails every month may deserve a fix for a different reason, namely the nuisance and cost. Keeping the questions separate avoids ranking an asset highly just because it has been annoying.
Common factors to score
Plants choose factors to suit their operation. Typical ones include:
- Safety and health. Could a failure injure people?
- Environment. Could it cause a release or a compliance problem?
- Production impact. Would output stop, slow down or continue through a backup?
- Quality impact. Could it cause off-specification product or scrap?
- Repair cost and time. Is the part easy to replace or does it have a long lead time?
- Redundancy. Is there a standby machine that can take over?
Redundancy deserves particular attention. A pair of identical pumps, one running and one standing by, usually lowers the production impact of a single failure. It does not remove the need to maintain the standby, since a spare that fails to start when called upon provides no protection.
A simple scoring method
Many teams use a small scale, for example 1 to 5 for each factor, with a short written description for each score so that different people rate the same way. The descriptions matter more than the numbers. “Score 5: failure stops the entire plant for more than a shift” is clearer than “high”.
Below is a hypothetical example, with made-up scores, for three assets. Each factor is scored 1 to 5 and the scores are added.
| Asset | Safety | Production | Repair time | Redundancy | Total |
|---|---|---|---|---|---|
| Main process feed pump (no standby) | 3 | 5 | 4 | 5 | 17 |
| Cooling water pump (with standby) | 2 | 2 | 3 | 1 | 8 |
| Break room exhaust fan | 1 | 1 | 1 | 1 | 4 |
Here redundancy is scored so that 5 means no backup and 1 means a full backup. The totals are 3 + 5 + 4 + 5 = 17, then 2 + 2 + 3 + 1 = 8, then 1 + 1 + 1 + 1 = 4. The numbers are illustrative only; a real plant would set its own factors and descriptions.
Some teams multiply scores or weight factors, giving safety extra weight so that a high safety score cannot be hidden by low scores elsewhere. A common safeguard is a rule that any asset with a serious safety score is treated as high priority regardless of its total.
Turning scores into groups
A long ranked list is hard to act on, so scores are usually grouped into a few bands such as high, medium and low. The bands then drive decisions:
- High criticality assets may get more thorough inspection routines, closer monitoring, written procedures and a stock of critical spares.
- Medium criticality assets usually follow standard preventive maintenance tasks on a fixed interval.
- Low criticality assets might be run to failure when that is safe and cheap, with repair done when needed.
Run-to-failure is a deliberate choice for low-consequence items, not neglect. It still needs a safety check, because something that looks minor may be linked to a hazard.
Who should take part
Maintenance staff know how machines fail, but operators know what happens to production when they stop, and safety, quality and planning staff each see different consequences. Scoring in a short group session usually produces better results than one person working alone from a spreadsheet. Disagreements are useful: they often reveal that people assume different things about backup equipment or process impact.
Common pitfalls
- Too many factors. A scheme with a dozen criteria becomes slow and inconsistent. Four to six is typical.
- Everything ends up “high”. If most assets rank at the top, the scale is not separating anything. Adjust the descriptions.
- Ranking at the wrong level. Rank at a level where decisions are made, usually the machine or system, not every individual bolt.
- Treating the result as permanent. Processes change, standbys are removed and new equipment arrives. Review rankings periodically and after major changes.
- Ignoring the result. If the ranking does not change how work is planned, it is only paperwork.
How the ranking is used
Once equipment is grouped, the ranking feeds other reliability work. Failure modes can be studied first for the highest bands. The choice of which items to hold as critical spares follows naturally, as do decisions about whether to add monitoring such as vibration checks. It also helps in planning a plant shutdown, because high-criticality assets are the ones people most want inspected while the plant is down.
Safety work on any equipment should follow the manufacturer’s instructions, the plant’s lockout and tagout style procedures, local regulations and trained, qualified personnel. This is general information, not engineering advice.