Production downtime dashboard: calculating MTTR, MTBF and OEE

What is MTTR and how is it calculated?

MTTR stands for Mean Time To Repair. It shows how long, on average, it takes to bring a piece of equipment back into service after it fails. The formula is simple:

MTTR = total repair time ÷ number of failures

Repair time runs from the moment the equipment stops to the moment it restarts, not just the technician's working time; waiting for the technician, finding the spare part and testing after the repair all count. That is why a high MTTR does not always mean weak technical skills; it is often a sign of missing spare parts or slow notification.

What is MTBF and what does it show?

MTBF stands for Mean Time Between Failures. It measures equipment reliability: the larger it is, the less often the equipment fails.

MTBF = operating time ÷ number of failures

Operating time is the time the equipment was scheduled to run minus the downtime caused by failures. Planned stops such as preventive maintenance or holidays are excluded. Together, MTTR and MTBF give the full picture: MTBF says how often failures happen, MTTR says how long each one lasts.

Worked example: MTTR and MTBF

Suppose press number 2 was scheduled for 600 hours in a month, failed 6 times, and the total downtime from those failures was 18 hours.

• Operating time = 600 − 18 = 582 hours
• MTBF = 582 ÷ 6 = 97 hours
• MTTR = 18 ÷ 6 = 3 hours
• Inherent availability = MTBF ÷ (MTBF + MTTR) = 97 ÷ 100 = 97 per cent

If maintenance cuts MTTR to 2 hours by keeping fast-moving spare parts on hand, availability rises to about 98 per cent and 6 hours of capacity are freed each month. If preventive maintenance reduces failures to 4, MTBF (with the same 3-hour MTTR) rises to 147 hours. A downtime dashboard should show these two improvement paths separately.

OEE: one number for breakdowns, slow running and scrap

OEE stands for Overall Equipment Effectiveness. It shows what share of planned production time was actually spent making good product at standard speed. OEE is the product of three ratios:

• Availability = run time ÷ planned production time (the effect of breakdowns and stops)
• Performance = (ideal cycle time × total count) ÷ run time (the effect of slow running and minor stops)
• Quality = good count ÷ total count (the effect of scrap and rework)
• OEE = availability × performance × quality

Worked example: OEE for one shift

Take an 8-hour shift (480 minutes) with a 30-minute planned break. During the shift there were 60 minutes of stops from breakdowns and a die change, the ideal cycle time is 1 minute per part, 351 parts were made and 14 of them were defective.

A quick check: 337 good parts × 1 minute ÷ 450 minutes also gives 74.9 per cent. Of every 100 planned minutes, about 75 went into making good product at standard speed. The real value of OEE is this breakdown: in this example the biggest loss is availability, so the priority is fewer breakdowns and faster die changes, not quality.

An OEE of 85 per cent is often quoted as world class, but many plants measure between 40 and 60 per cent when they start. Comparing a line with its own previous months is more useful than comparing it with a general benchmark.

Recording downtime correctly: the base of all three KPIs

No dashboard is better than the data it receives. The problems I see most in plant downtime data: the operator, the shift lead and maintenance each log the same failure; short stops of a few minutes are never recorded; the reason is typed as free text, so "electrical fault", "power problem" and "power cut" count as three different causes.

A good downtime form has these fields: line and equipment (from a fixed equipment tree), start and end time, stop type (planned or unplanned), a reason from a standard list and an optional note. With start and end times, the dashboard can merge overlapping stops. If one line has two records, 10:00 to 10:40 and 10:20 to 11:00, its net downtime is 60 minutes, not 80.

What a good production downtime dashboard contains

The production line downtime dashboard I built for a real plant has these parts (see the walkthrough with made-up data in the production line downtime dashboard portfolio piece):

• A downtime timeline for each day and shift, showing which hours the stops cluster in.
• An equipment tree from line to machine to component, with downtime and failure count for each.
• A Pareto chart of stop reasons to pick the top three problems.
• MTTR, MTBF and availability cards per equipment, with their monthly trend.
• A monthly calendar that colours the days with the most downtime.
• OEE by line, shift and product.

The dashboard is built in Power BI and refreshes daily from the downtime log. To put a price on every hour of downtime, connect it to the product costing model: lost capacity × the fixed overhead per hour.

From dashboard to improvement

A dashboard alone does not reduce downtime; the routine around it does. I suggest a short weekly meeting between production and maintenance: look at the week's downtime Pareto, pick the top three causes, assign an action and an owner to each, and measure the effect in the same dashboard the following week. Within a few months this simple routine usually frees more capacity than any large investment. The service is described on the production and downtime dashboard page, and you can try an interactive production dashboard sample.

Frequently asked questions