Calculating MTBF can be fairly simple or really complex. We’ve included links to some resources, including definitions, downloadable Excel calculators, and online calculators. Some are simpler to use than others. But that’s not surprising. Different experts describe how to calculate MTBF and its colleagues MTTR and MDT with varying degrees of complexity.
If you are a statistics nerd and love digging in to the formulas then you’ll find plenty to love online, by way of some of the links below. If you want to use MTBF for a very specific, maybe even a crude or rough calculation, you’ll also find some simpler tools here. And, if you just want to know what the heck all of these acronyms mean and why you should care, read on. We’ll try to help.
What Is MTBF?
MTBF, or Mean Time Between Failure, tells you the average amount of uptime you are getting from a machine or system. As you’ll learn if you read through some of the pages linked below, there are a lot of factors involved in calculating a truly accurate MTBF. But, in its most simple sense, MTBF is the average amount of time a machine or system is up and running between failures.
Ferdna Andjohn2000 [Public domain], via Wikimedia Commons
What MTBF Isn’t
MTBF is about unplanned downtime or failures for repairable systems. It doesn’t include planned shutdowns or preventive maintenance, nor does it work for unrepairable systems.
The basic MTBF result is arrived at by adding up all of the amounts of time that the machine has been operational during some time period. Let’s assume you will count unplanned failures that occur between your planned shutdowns. If you shutdown once a year, then you will keep track of the amount of time the machine is operational between each failure.
If you’re using a CMMS, it should make it easy for you to track both downtime and uptime for any machine or system. If you’re not, why not?
Ok, if you’re not, then we hope you’re keeping a log, writing down the time and date when a machine fails or breaks down, and then the time and date when that machine comes back online after being repaired.
You should end up with a list of dates and times for each machine. You can determine the amount of time between each of those, coming up with numbers of hours of downtime and uptime for each failure and repair occurrence.
MTBF calculation depends on your situation
MTBF calculation depends on the situation in your environment. For instance, do you run your machines 24/7? Or do you only run two shifts a day? Do you have multiple machines of the same age and make running the same process and products?
The answers to these and other questions will change the way you calculate MTBF, as you will find by doing some research at the pages we link to below.
Why Calculate MTBF at All?
As funny as that sounds, it’s actually a good question. MTBF is often one of the main statistics used if your company has implemented a predictive maintenance program. MTBF helps maintenance engineers predict when a machine or system will fail. Using those statistics, they can implement a system of part replacement and planned maintenance, in the effort to eliminate as much unplanned downtime as possible.
But, even as a Maintenance Manager, you can use these statistics to help you anticipate problems and understand your equipment.
Watching MTBF Trends
If all your production equipment is brand-new, you will spend more of your time collecting data for future reference. Just because things aren’t breaking left and right, don’t get complacent. The data you collect today may save you a lot of time, effort, and money months or even years down the road.
Now, imagine you’ve been tracking MTBF for a specific piece of equipment over the last three years. You’ve had your techs run through all of the planned maintenance. Now and then, the machine has failed and you’ve had to repair and replace some parts. Just thinking back to all of the maintenance and repairs on that machine, you may develop some sense or idea about whether or not that machine is a big risk. Unfortunately, we all have a lot of machines and systems to think about. Using our gut feelings isn’t always the most accurate way to approach maintenance.
Instead, reviewing three years worth of MTBF stats, you might see something interesting by comparing the change in numbers, or better yet, a graph of the MTBF numbers. Let’s say that In the 2nd year of operation, MTBF decreased slightly, month-over-month, but not enough to raise any red flags. In the 3rd year of operation, MTBF decreased by about 1.5% month-over-month. While this still might be a very slight change in absolute terms, it’s troubling. The velocity of the change in MTBF is something you should not ignore, even if the total amount of uptime decreased by only 3 or 4 hours over the course of a year. The fact that uptime is decreasing steadily over time should alert you that something is amiss.
Troubleshooting MTBF Trends
Troubleshooting MTBF trends is much like troubleshooting machine problems. Only you need to consider more unrelated variables: equipment quality, equipment location, operator skill, maintenance technician skills, changes in equipment processes or materials, and others.
Looking back at the MTBF numbers, can you identify when the MTBF began to decrease steadily? A graph of the numbers can be especially helpful. You can easily chart the MTBF readings over time in Excel or Google Sheets.
Locating the beginning of the MTBF decrease, consider the following:
- What changed around the time the MTBF began to decrease?
- Was the equipment put to a different use?
- Did different operators take over running the equipment?
- Did different technicians take over maintaining the equipment?
- Did you change vendors for parts or outsourcing maintenance on the equipment?
- Was there a change in the physical environment in which the equipment is located?
- Has your training program changed?
- Has your compliance program changed?
With new equipment, it can actually be more difficult to tell if this is just the equipment’s natural deterioration or if the decrease in MTBF is due to external factors. When tracking an older, well-known piece of equipment, you will likely have a more detailed history of failures, maintenance, and uptime. Negative trends may be easier to identify.
Talk to Your Equipment Vendor
The vendor who sold you that piece of equipment also gave you MTBF standards, or other reliability measures for the equipment. You can check your numbers against the ones provided by the manufacturer to see if your numbers are outliers. You should also consider comparing numbers with other equipment users. If your plant is part of a large conglomerate that has similar plants operating elsewhere, you should be able to get data from your colleagues in those other plants. If you’re using a corporate-wide CMMS, you should be able to find that data, although you may need to filter and format it for your own purposes.
If you are a solo shop, talk to the vendor. If your numbers are way out of scope, the vendor will want to figure out the problem as much as you do.
We’ve just barely touched on MTBF, not to mention the many other useful concepts and acronyms, but we hope you see how you can use even a simplistic set of these calculations to help anticipate and plan for equipment maintenance or repair. Explore the links below to see if there are resources you might find useful. We’ve tried to include only those that are fairly straightforward to understand and use, and of course, are free. However, inclusion on this list does not mean we take responsibility for their use or results. We have no relationship with any of these sites and do not benefit from their inclusion here.
FDA: Reliability of Manufactured Products
Mean Time Between Failure: Explanation and Standards
Research Brings Results – Defining Mean Time Between Pump Failures
Reliability Block Diagram
Reliability Analytics Toolkit
ALD Service Reliability Software
Free MTBF Calculator from SoHaR