How is the criticality score calculated?

Consequence times likelihood. With the weighted average method, consequence is each 1 to 5 score multiplied by its weight, summed and divided by the sum of the weights, rounded to 0.1. The example reactor agitator scores 3, 3, 5, 4 and 4 for safety, environment, production, quality and repair cost; with weights of 30, 20, 30, 10 and 10 percent its consequence is 3.8, and at likelihood 4 its criticality is 3.8 x 4 = 15.2: rank 1, and class A at the example threshold of 12.

Why is a relief system that almost never fails ranked class A?

Because of the safety and environment rule on Setup. With the rule set to Yes, any asset scored 5 for safety or environment is class A however unlikely its failure. The example reactor relief system scores 5 for both, its consequence is 3.4 and its likelihood 1, so its criticality is only 3.4 and it ranks 27th of 29, yet it is class A; without the rule it would be class C. The risk map shading follows the score thresholds only, so it does not show this rule.

Should I use the weighted average or the highest score method?

A weighted average can hide one severe consequence: with the example weights, an asset scored 5 for safety and 1 for everything else averages 2.2. The highest score method takes the worst of the five, 5 in that case, so nothing is averaged away, but it ignores the weights and gives more ties, because every criticality score is then the product of two whole numbers. Choose with your team before scoring and keep the method fixed so rankings stay comparable. The Verify sheet tests the weighted average method; switching method changes your results, not the self-check.

Where do the class thresholds, weights and scales come from?

From your team. Class A from a score of 12, class B from 6, the weights and the scale wording are example choices for the fictional plant, not values from a standard; with them the example register has 5 class A, 18 class B and 6 class C assets. The Method sheet cites ISO 55000, IEC 60812, SAE JA1011 and NORSOK Z-008 for the approach, but the scales are the workbook's own wording, not tables from those standards. Setup warns if the weights do not add up to 100 percent or the class B threshold is above class A.

How is the downtime cost at stake calculated, and does it change the class?

Cost per failure is hours per failure times lost output per hour plus repair cost per failure, and annual downtime cost is failures per year times cost per failure. The example reactor agitator fails 3 times a year for 10 hours at 4,500 an hour with 6,000 of repair: 10 x 4,500 + 6,000 = 51,000 per failure and 153,000 a year. These are the Downtime Cost per Event and Annual Downtime Cost calculators' formulas, and the Verify sheet checks the calculator's own example (68,000 per event). The cost is an estimate shown beside the ranking; it never changes the class, and it leaves out knock-on losses such as missed orders unless you add them to the repair cost.

Can I rank more assets or areas than the example?

Yes. The Asset Register and the Setup lists are Excel tables that grow as you type, and every new asset is scored, classed and ranked; the class counts, cost tiles, by-class roll-up and risk map cover the whole register. The Dashboard lists the top 15 assets, and its By area block is sized to the five example areas: if you add areas, extend the block (see Method), and the Dashboard shows a warning while the Areas list is longer than the block.

Maintenance and Reliability

Asset Criticality Ranking Excel Template

Score every asset on what its failure would do and how likely that failure is, rank the register into A, B and C classes, and see on one dashboard where the risk and the downtime money sit.

Overview

This asset criticality ranking is a working Excel tool for reliability and maintenance engineers, maintenance managers, asset owners and plant managers who have to decide where maintenance effort, condition monitoring and critical spares go first. It ships with 29 example assets in five areas of a fictional specialty-chemicals plant, each scored by hand, so the dashboard is populated the moment you open it: a reactor agitator and a decanter centrifuge reach class A on their score, while a relief system, a fire water pump and a vent gas scrubber that rarely fail are class A because their failure could be a safety or environmental event.

Each row is one asset. You score it 1 to 5 on five consequences of failure (safety, environment, production, quality and repair cost) and 1 to 5 on likelihood, using the scale in words on Setup. The workbook combines the five consequences into one score, either as a weighted average with your weights or as the highest of the five, multiplies it by likelihood for a criticality score from 1 to 25, and turns the score into class A, B or C at the thresholds your team sets. Every scored asset gets a unique rank. Optional failure history gives an estimated cost per failure and annual downtime cost, the same formulas as the Downtime Cost per Event and Annual Downtime Cost calculators on this site, so the ranking can be argued with in money; the cost never changes the class.

The dashboard shows the class counts, the highest-ranked asset and the downtime cost at stake, the top 15 assets, the classes by area and a 5 x 5 risk map that counts assets by likelihood and consequence, each cell shaded by the class its score falls in. In the example, class A holds 5 of the 29 assets and 27 percent of the estimated downtime cost, and class B holds 69 percent, which is the kind of result to take back to the scoring team. The scales, weights and thresholds are example choices, not values from a standard: agree your own before you score and keep them fixed while you do.

What this template includes

Ranking rules on Setup: consequence method (weighted average or highest score), five criterion weights with a check that they add up to 100%, class A and B score thresholds and a rule that a safety or environment 5 is class A
A scoring scale in words for safety, environment, production, quality, repair cost and likelihood, and the maintenance, spares and failure follow-up each class suggests (example wording to rewrite for your site)
Asset Register table: asset ID, asset, area and a 1 to 5 score for each consequence and for likelihood, with an area dropdown and whole-number validation on every score
Per asset: consequence (1-5), criticality score (1-25) with an in-cell bar, class A, B or C and a unique rank, with equal scores ordered by asset ID
Optional failure history (failures per year, hours per failure, lost output per hour, repair cost per failure) for an estimated cost per failure and annual downtime cost
A Check column that names problems in words: a missing score, asset or area, a duplicate asset ID, a score outside 1 to 5, failures with no hours per failure
Dashboard tiles: assets ranked, class A, B and C counts, the highest-ranked asset, class A and total downtime cost per year, and rows with a Check message
A top 15 list and bar chart, a chart of class A, B and C assets by area, a 5 x 5 risk map shaded by class zone, and roll-ups by class and by area with each one's share of the downtime cost
Method sheet with every formula, a Verify sheet that re-checks them against an independent reference (Self-check: PASS), and a CSV with live formulas of the Asset Register; the dashboard, charts, dropdowns and self-check are in the .xlsx

Suggested use case

Use it to decide which assets get condition monitoring, preventive maintenance and critical spares first, to pick the assets for an RCM or FMECA study, or to show why the maintenance plan treats assets differently.

How to use it

On Setup, agree the ranking rules with operations, safety and maintenance: consequence method, criterion weights, the class A and B thresholds and whether a safety or environment 5 makes an asset class A.
Rewrite the example scoring scale and the strategy for each class in words that fit your site, and replace the example areas.
On Asset Register, add one row per asset and score it 1 to 5 on safety, environment, production, quality, repair cost and likelihood; add failures per year, hours per failure, lost output per hour and repair cost per failure where you have them.
Fix anything the Check column reports.
Open the Dashboard for the class counts, the top 15 and the risk map, then filter the Asset Register's Class column for A and work every class A asset first, including those that are class A only because of the safety and environment rule; use the downtime cost by class and area to test the ranking against what the plant knows.

Frequently Asked Questions

How is the criticality score calculated?: Consequence times likelihood. With the weighted average method, consequence is each 1 to 5 score multiplied by its weight, summed and divided by the sum of the weights, rounded to 0.1. The example reactor agitator scores 3, 3, 5, 4 and 4 for safety, environment, production, quality and repair cost; with weights of 30, 20, 30, 10 and 10 percent its consequence is 3.8, and at likelihood 4 its criticality is 3.8 x 4 = 15.2: rank 1, and class A at the example threshold of 12.
Why is a relief system that almost never fails ranked class A?: Because of the safety and environment rule on Setup. With the rule set to Yes, any asset scored 5 for safety or environment is class A however unlikely its failure. The example reactor relief system scores 5 for both, its consequence is 3.4 and its likelihood 1, so its criticality is only 3.4 and it ranks 27th of 29, yet it is class A; without the rule it would be class C. The risk map shading follows the score thresholds only, so it does not show this rule.
Should I use the weighted average or the highest score method?: A weighted average can hide one severe consequence: with the example weights, an asset scored 5 for safety and 1 for everything else averages 2.2. The highest score method takes the worst of the five, 5 in that case, so nothing is averaged away, but it ignores the weights and gives more ties, because every criticality score is then the product of two whole numbers. Choose with your team before scoring and keep the method fixed so rankings stay comparable. The Verify sheet tests the weighted average method; switching method changes your results, not the self-check.
Where do the class thresholds, weights and scales come from?: From your team. Class A from a score of 12, class B from 6, the weights and the scale wording are example choices for the fictional plant, not values from a standard; with them the example register has 5 class A, 18 class B and 6 class C assets. The Method sheet cites ISO 55000, IEC 60812, SAE JA1011 and NORSOK Z-008 for the approach, but the scales are the workbook's own wording, not tables from those standards. Setup warns if the weights do not add up to 100 percent or the class B threshold is above class A.
How is the downtime cost at stake calculated, and does it change the class?: Cost per failure is hours per failure times lost output per hour plus repair cost per failure, and annual downtime cost is failures per year times cost per failure. The example reactor agitator fails 3 times a year for 10 hours at 4,500 an hour with 6,000 of repair: 10 x 4,500 + 6,000 = 51,000 per failure and 153,000 a year. These are the Downtime Cost per Event and Annual Downtime Cost calculators' formulas, and the Verify sheet checks the calculator's own example (68,000 per event). The cost is an estimate shown beside the ranking; it never changes the class, and it leaves out knock-on losses such as missed orders unless you add them to the repair cost.
Can I rank more assets or areas than the example?: Yes. The Asset Register and the Setup lists are Excel tables that grow as you type, and every new asset is scored, classed and ranked; the class counts, cost tiles, by-class roll-up and risk map cover the whole register. The Dashboard lists the top 15 assets, and its By area block is sized to the five example areas: if you add areas, extend the block (see Method), and the Dashboard shows a warning while the Areas list is longer than the block.