An asset criticality ranking matrix scores every asset by the risk it carries, criticality = probability of failure × consequence of failure, across factors such as safety, environment, production, quality, and cost. Assets are then sorted into tiers so maintenance effort, spares, and monitoring follow risk instead of habit or the loudest complaint.
Every plant has finite maintenance hours, and criticality ranking is how you decide where they go. Without it, effort flows to whatever broke last or whoever shouts loudest, and the quiet machine that will shut the whole line down next month gets the same routine as a spare fan. This guide is the how-to: the factors that make up a criticality score, a weighting template you can adapt, the workshop that produces the scores, and how to turn them into tiers that actually change what your crew does.
What is an asset criticality ranking matrix?
An asset criticality ranking matrix is a structured, semi-quantitative method for rating the relative importance of every physical asset by combining how likely it is to fail with how bad the consequences would be if it did. The output is a single criticality score per asset and a ranked list, usually grouped into tiers, that tells you which assets deserve the most reliability attention.
The core equation is risk: criticality = probability of failure × consequence of failure. An asset that fails often but harmlessly and an asset that almost never fails but would be catastrophic can land at the same score, and both may matter less than the asset that fails fairly often and hurts badly when it does. Multiplying the two dimensions is what surfaces that middle case, which pure failure-frequency or pure consequence thinking both miss.
Why rank asset criticality?
Because maintenance strategy should be assigned per asset, and criticality is the input that decides it. The ranking tells you which machines justify condition monitoring and predictive investment, which get a tight preventive schedule, and which can safely run to failure. It also drives spares stocking, PM interval selection, inspection frequency, and where you spend engineering time on root-cause work.
Just as importantly, it settles arguments. A documented criticality ranking replaces "everything is critical", the reflexive answer from every department about its own equipment, with a defensible, shared list. That single artifact underpins the whole equipment reliability program: it is the map that tells you where on the maturity ladder each asset should sit, and it is why criticality ranking is usually one of the first things a serious reliability effort produces.
What factors go into a criticality score?
Consequence is not one thing, so a good matrix breaks it into weighted factors and scores each asset on all of them. The industry-standard approach is semi-quantitative weighted scoring, used in frameworks like the SMRP body of knowledge, ISO 31000 risk management, and API 580 risk-based inspection. A typical factor set, with example weights that sum to 100%, looks like this, treat it as a template to adapt to your plant's real priorities.
| Factor | What it measures | Example weight |
|---|---|---|
| Safety | Risk of injury to people if the asset fails | 25% |
| Environment | Risk of spill, release, or regulatory breach | 15% |
| Production impact | Lost output and whether failure stops a line or the plant | 25% |
| Quality / compliance | Effect on product quality or regulatory requirements | 10% |
| Repair cost | Cost to repair or replace, including collateral damage | 10% |
| Spares & lead time | Availability and lead time of critical parts | 10% |
| Failure frequency | How often it fails, from history or MTBF (the probability side) | 5% |
Score each factor on a simple scale, 1 to 5 is common, with a clear rubric for what each level means, multiply by the weight, and sum to a weighted consequence score. Combine that with the probability rating to place the asset on the matrix. The rubric matters more than the math: if "5" on safety means "potential fatality" and everyone scores against the same words, the numbers become comparable across the plant.
How do you run a criticality ranking workshop?
Criticality scoring is a cross-functional exercise, not a solo spreadsheet. The consequences span safety, operations, quality, and cost, so the people who understand each belong in the room. Run it like this.
- Build the asset list from your hierarchy. Start from a clean asset hierarchy so you rank at a consistent level, usually the equipment-unit level, and nothing is missed or double-counted.
- Agree the factors and weights. Decide the consequence factors and their weights before scoring any asset. Doing this first prevents the weights from being reverse-engineered to justify favorite machines.
- Write the scoring rubric. Define what each level (1 to 5) means for every factor in plain words, so "high production impact" means the same thing to everyone.
- Assemble the right people. Operations, maintenance, reliability, safety, and quality all score together. Criticality set by maintenance alone will miss consequences the other functions see.
- Score each asset across all factors. Work through the list, rate probability and each consequence factor, and let the workshop debate the outliers, the debate is where tribal knowledge surfaces.
- Calculate and rank. Compute the weighted consequence, combine with probability, and produce the ranked list and matrix placement.
- Assign tiers and actions. Group assets into tiers and attach a maintenance strategy to each tier, so the ranking changes what actually happens on the floor.
How do you turn scores into tiers?
A ranked list is only useful if it drives action, so group the assets into a few tiers and attach a maintenance strategy to each. A common scheme is three tiers, sometimes with a fourth for safety-critical outliers:
- Tier A (critical). High criticality, often the top 10–20% of assets that carry most of the risk. These earn condition monitoring or condition-based maintenance predictive investment where justified, tight PM intervals, and critical spares held on the shelf.
- Tier B (important). Moderate criticality. A disciplined preventive maintenance schedule and standard spares, reviewed periodically.
- Tier C (low). Low criticality, cheap, redundant, or easily replaced assets. Run-to-failure or minimal PM is the deliberate, correct choice, freeing hours for the tiers that matter.
The tiering usually reveals a familiar pattern: a small share of assets carries the majority of the risk. That concentration is exactly why ranking pays off, it lets you pour disproportionate attention onto the vital few and stop over-maintaining the trivial many. Feeding those tier assignments into maintenance planning and scheduling is what turns the analysis into a changed weekly schedule.
How does criticality ranking relate to FMEA?
Criticality ranking and FMEA (failure mode and effects analysis) work at different resolutions and feed each other. Criticality ranking operates at the asset level and answers "which machines deserve the most attention?" FMEA drills into an individual asset and answers "how exactly does this machine fail, and which failure modes matter most?" You do not FMEA every asset, that would take forever, so criticality ranking is what tells you which assets are worth the deeper FMEA effort.
The two share the same risk logic. FMEA's risk priority number multiplies severity, occurrence, and detectability at the failure-mode level, mirroring the probability-times-consequence thinking at the asset level. Run criticality ranking first to find the vital few, then FMEA those assets to design the specific maintenance tasks, condition-monitoring points, and spares that protect them. Ranking sets the priorities; FMEA turns a priority into a concrete plan.
How often should you review criticality?
Criticality is not set once and forgotten. Review it when the plant changes, a new line, a debottlenecking that shifts which machine is the constraint, a redundancy added or removed, and on a periodic cycle, commonly annually, to catch drift. An asset that was a bottleneck last year may not be this year, and its ranking should follow.
Failure history is the other trigger. If an asset keeps failing in ways the ranking did not anticipate, its probability rating was wrong and the score needs updating. This is where good downtime and work-order data closes the loop: the same records that feed availability and MTBF also validate whether your probability ratings match reality. Ranking on gut feel is a fine start; ranking that gets corrected by data is what keeps it trustworthy.
Where criticality ranking fits your reliability program
Criticality ranking is the decision layer that makes every other maintenance investment rational. It tells you where to put sensors, which spares to stock, how tight to set PM intervals, and where run-to-failure is smart rather than lazy. Done in a cross-functional workshop against a clear rubric, it produces a shared, defensible list that outlasts any one person's opinion, and it belongs upstream of almost everything else you do.
Keeping it current is the hard part, because it depends on live failure and downtime data most plants have scattered across systems. That is the layer machine-monitoring platforms like Harmony provide, connecting machine signals, downtime reasons, work-order history, and spares status into one operational view, so your probability ratings stay grounded in what is actually failing and your tier assignments drive real, scheduled work. It layers onto the CMMS and systems you already run, with no rip-and-replace. See how the platform works or read the CLS case study.