What Is Asset Uptime?

Asset uptime is the percentage of scheduled operating time during which an asset is actually available and working. It is the simplest answer to the question “can we rely on this thing?” - and the headline number that planned and unplanned maintenance decisions are ultimately judged against.

This guide covers the formula and a worked example, how uptime relates to availability and to the reliability metrics MTBF and MTTR, what to count as downtime, what an hour of it costs, the IT “nines” meaning, and realistic targets.

What you will learn

• The uptime formula
• The metrics behind uptime: MTBF, MTTR and OEE
• Uptime vs availability vs reliability
• Planned vs unplanned downtime: what to count
• What an hour of downtime costs
• Uptime in IT vs uptime for equipment
• What drags uptime down
• Realistic uptime targets
• Tracking uptime in practice
• FAQ

The uptime formula

Uptime = (scheduled operating time - downtime) ÷ scheduled operating time × 100.

A worked example: a workshop compressor is scheduled to run 40 hours in a week. A blown hose stops it for 2 hours on Wednesday. It ran 38 of its 40 scheduled hours, so uptime for the week is 38 ÷ 40 = 95%.

The same formula scales to any window. Take a machine scheduled around the clock - 720 hours in a 30-day month - that loses 18 hours to assorted stoppages: (720 - 18) ÷ 720 = 97.5%. Over a full year of 8,760 hours, hitting 99% leaves a budget of roughly 88 hours of downtime. Quoting downtime in hours per period is often clearer than the percentage alone, because a single decimal place hides a lot of lost production.

The arithmetic is trivial; the definitions are where teams disagree. Decide up front what “scheduled time” means (24/7, or working hours only?) and whether planned servicing counts as downtime. The common convention is to exclude planned maintenance windows from scheduled time, so the figure isolates unexpected failures - the stoppages you can actually do something about.

The metrics behind uptime: MTBF, MTTR and OEE

A single uptime percentage tells you how much an asset was down, but not why. Two reliability metrics break that number into its two real drivers:

• MTBF (mean time between failures) = total run time ÷ number of failures. It answers “how often does it break?”. A machine that runs 1,000 hours across 5 failures has an MTBF of 200 hours.
• MTTR (mean time to repair) = total repair time ÷ number of failures. It answers “how long does it take to get going again?”. If those 5 repairs took 25 hours in total, MTTR is 5 hours.

The two combine into the availability identity that competitors use to connect uptime and availability formally:

Availability = MTBF ÷ (MTBF + MTTR).

With the numbers above, that is 200 ÷ (200 + 5) = 97.6%. The value of splitting it this way is that it points at the fix. A poor figure caused by a short MTBF is a reliability problem - chase it with preventive maintenance and better parts. A poor figure caused by a long MTTR is a response problem - chase it with spare-parts readiness and faster corrective maintenance.

OEE (overall equipment effectiveness) widens the lens further. Where uptime only counts full stoppages, OEE multiplies availability by performance (slow cycles and minor stops) and quality (scrap and rework). An asset can be 98% “up” yet still bleed output through running below rated speed - which uptime alone will never reveal.

Uptime vs availability vs reliability

The three words get used interchangeably, but they answer different questions:

A few consequences fall out of that table:

• Uptime measures performance against the hours you planned to use the asset.
• Availability exposes assets that are “up” but unusable - out of calibration, missing a part, or simply nowhere to be found.
• Reliability is why two assets can post identical uptime yet feel completely different: one had a single long outage, the other a stoppage every shift. The second is far less reliable, and far more disruptive.

The distinction matters beyond machinery. Safety kit such as fall protection equipment that has missed its inspection date is effectively down even though nothing is broken - it cannot lawfully be used. That is the everyday answer to “why is my equipment available but not reliable?”: availability is a snapshot, reliability is a track record.

Planned vs unplanned downtime: what to count

Not all downtime is a failure, and treating it as one distorts the number. Two categories matter:

• Planned downtime - scheduled servicing, calibration, changeovers and upgrades. You chose the timing, so it is usually excluded from scheduled operating time rather than counted as a fault. MTBF deliberately ignores it for the same reason.
• Unplanned downtime - breakdowns, awaiting parts, operator-found defects. This is the figure you act on, because it represents capacity you expected to have and lost.

Drawing the line cleanly is what makes uptime comparable month to month. If a planned service window quietly counts as downtime one quarter and not the next, the trend becomes meaningless. The cleaner split between scheduled work and surprises is exactly what the planned vs unplanned maintenance distinction exists to formalise, and it feeds directly into the wider downtime picture.

What an hour of downtime costs

“Is uptime worth chasing?” only has an answer once you can price an hour of downtime. There is no universal figure to borrow, but there is a reliable method:

Downtime cost per hour = lost output value + idle labour + recovery cost.

• Lost output value - the revenue or production an hour of running would have generated, net of materials you did not consume.
• Idle labour - people paid to stand by while the asset is down, including those downstream of it.
• Recovery cost - overtime, expedited parts, or weekend shifts spent catching up.

Calculate this once for your own assets and the business case writes itself. An asset whose downtime costs a few euros an hour does not justify a costly maintenance programme; one that idles a whole line at hundreds of euros an hour easily does. This figure - not a benchmark copied from another company - is what should set your target.

Uptime in IT vs uptime for equipment

Two very different audiences search for “uptime”, and the word means slightly different things to each.

In IT and service-level agreements, uptime is measured against 24/7 clock time and quoted in “nines”: 99.9% (“three nines”) allows about 8.8 hours of downtime a year, or roughly 43 minutes a month; 99.99% tightens that to under an hour a year. These commitments are usually written into a service-level agreement with penalties attached.

For physical equipment, the “nines” language rarely fits. Most machines are measured against scheduled hours, not the full calendar, and a workshop tool does not need - or cost-justify - four nines of availability. For physical kit it is almost always clearer to state a target as “no more than X hours of unplanned downtime per month” than to argue about decimal places. The bridge between the two worlds is the formula: both are operating time over expected time; only the denominator and the stakes differ.

What drags uptime down

The usual suspects are unglamorous, and each maps to MTBF or MTTR:

• Skipped servicing lets small wear become a breakdown - it shortens MTBF.
• Awaiting parts stalls repairs because nobody stocked the spare - it lengthens MTTR.
• Mystery and administrative downtime - the asset is fine but checked out to someone unknown, sitting in the wrong van, or waiting on a ticket no one is watching.

For shared kit such as HVAC tools, time spent hunting for an instrument is downtime in every way that matters, even though the tool itself never failed. Naming the cause is half the fix: once a stoppage is filed against the right category, the biggest lever becomes obvious.

Realistic uptime targets

IT teams talk in “nines”, but physical equipment rarely justifies that language. As market context, maintenance practitioners often treat around 90% as a floor for general equipment and aim for 95-99% on critical assets - useful orientation, but not a target to adopt blindly. A sensible approach for everyday business assets is to baseline first: log every stoppage for a quarter, calculate the actual figure, then set a target one notch better and aim at the biggest single cause. A preventive maintenance schedule for the worst offenders, paired with the cost-per-hour number above, usually moves the figure more than any tooling change.

Tracking uptime in practice

You cannot improve a number you never record, and downtime rarely records itself. The habit that works is small: every stoppage becomes a logged event against the asset - when it failed, why, and when it came back. In AMPthilly, each stoppage is a service-desk ticket tied to the asset record, with a category (damage, missing, needs maintenance, needs replacement), a status as it moves through the queue (in review, in progress, awaiting parts, resolved), and comment threads plus attached repair invoices. That ticket history stays on the asset permanently, so the downtime story for any item is already written when you come to count it. Review the log monthly, and the assets that deserve a maintenance plan - or retirement - identify themselves.

FAQ

How do you calculate asset uptime?

Divide the time the asset was actually working by the time it was scheduled to work, then multiply by 100. A machine scheduled for 40 hours that lost 2 hours to a breakdown ran 38 of 40 hours, giving 95% uptime. The key decision is what counts against the asset - most teams exclude planned servicing windows from scheduled time and count only unexpected stoppages as downtime.

What is the difference between uptime and availability?

In everyday use they overlap, but uptime usually asks “was it running during scheduled hours?”, while availability asks “could it have run whenever it was needed?”. An asset in a locked store with a flat battery has technically had no downtime, yet it was not available. Availability also tends to fold in planned maintenance windows, whereas uptime calculations often exclude them.

What is a good uptime percentage for equipment?

There is no universal target - a hospital generator and a shared office projector justify very different standards. Rather than chasing a borrowed benchmark, measure your own baseline for a few months, then set a target that reflects what an hour of downtime actually costs you. For most teams the trend matters more than the number: rising unplanned stoppages are the real warning sign.

What is MTBF and how does it relate to uptime?

MTBF (mean time between failures) is total run time divided by the number of failures - it measures how often an asset breaks. Paired with MTTR (mean time to repair), it explains a high uptime figure: availability equals MTBF divided by (MTBF + MTTR). A long MTBF and a short MTTR both push uptime up, so improving either one moves the number.

What does 99.9% uptime mean?

99.9% uptime, often called “three nines”, allows roughly 8.8 hours of downtime across a full year, or about 43 minutes a month. The “nines” shorthand comes from IT and SLAs measured against 24/7 clock time; for physical equipment measured against scheduled hours it is usually clearer to quote downtime in hours per period than to chase extra nines.

How do you calculate uptime over a month or a year?

Use the same formula over the longer window. For a machine scheduled 720 hours in a month (24 hours x 30 days) that lost 18 hours to stoppages, uptime is (720 - 18) / 720 = 97.5%. Over a 8,760-hour year, 99% uptime leaves about 88 hours of downtime. Always state the schedule you are measuring against, since “scheduled hours” and “calendar hours” give very different percentages.

Tools that make this easier

AMPthilly keeps the raw material for an uptime number where it belongs - on the asset. Every issue report and repair is a service-desk ticket tied to the item, with statuses, comment threads and attached invoices, and the full audit history of checkouts, returns and status changes stays on the record permanently. Scan an asset’s QR label with a phone camera (no app to install) to open its profile and report a stoppage on the spot. The free plan covers 3 users and 25 assets with no card required, so you can start logging downtime before you formalise a target.

The takeaway

Asset uptime is operating time over scheduled time, expressed as a percentage - simple to compute, but only meaningful once you have settled what counts as scheduled time and what counts as downtime. Break the number into MTBF and MTTR to see whether you have a reliability problem or a response problem, price an hour of downtime to set a target worth chasing, and log every stoppage against the asset so the trend - the part that actually warns you - is always there to read.

Related terms

• Downtime - the time you subtract, and the figure to drive down
• Planned vs Unplanned Maintenance - the split that decides whether downtime surprises you
• Corrective Maintenance - the repairs that determine your MTTR
• Preventive Maintenance - scheduled work that trades small planned stops for big unplanned ones
• Equipment Servicing - the routine care that keeps uptime high
• Spare Parts Management - stocking the parts that turn long outages into short ones
• Service Level Agreement (SLA) - the contract form uptime commitments usually take