About a year ago, I started thinking about what I needed to do in order to make my business better, and it occurred to me that there were a few issues that I needed to focus on. Since I wasn't sure where to start, I decided to hire a business consultant, and it made a world of difference. Within a few short weeks, I had uncovered a trail of issues that had to be fixed, and making the changes really helped to turn things around. This blog is all about working with a great consulting firm and improving the quality of your company.
Many organizations follow what is fundamentally a break-fix cycle when it comes to maintaining equipment and systems. Unfortunately, this approach can prove to be expensive, especially if you're dealing with something that may fail catastrophically.
One way to address the situation is to figure out what the mean time between failures is likely to be for key systems. A good MTBF calculation can help a business reduce accidents, avoid expenses, and improve quality. Let's look at how you can end the break-fix cycle by adopting an MTBF prediction model.
Data Collection
Foremost, you need to have quality data before you can even think about MTBF prediction and calculation. This means you'll need to gather relevant data about failures. More importantly, the data needs to be robust and accurate across a significant period.
Suppose your company owns a large commercial building with several elevators. The business needs to reduce expenses tied to elevator repairs. You would have operators and technicians log all run times and failures over several months.
Notably, you should taint the data collection process by feeding biases into it. There is no need to tell people in the building that you're doing a study, for example. Not even the technicians who repair the systems need to know. Simply collect the data in as undisturbed and neutral of an environment as possible.
Preparing the Data
It is also necessary to prepare the data for analysis. There might be outliers. An electrical company preparing for an MTBF calculation may identify impossible readings, for example. You should identify outliers and track them down. If the real-world evidence indicates the outlier is legitimate, you should keep it in the data. However, if it is the product of instrument or transcription errors, you can usually eliminate it from the analysis.
Calculating the MTBF
Working with the data, a consultant can help you calculate what the MTBF is likely to be. The mean time between failures, in this case, means the average. If you have five failures over 8 weeks, for example, then the MTBF is 11.2 days.
Bear in mind that the calculation may be more complex. For example, a mining company might use a calculation that factors in weather and worker stress to think about a range of scenarios.
Application
With a calculation in hand, you can now make an MTBF prediction. Companies often apply these predictions through software programs, spreadsheets, or guidelines. A bus company, for example, might order maintenance checks on vehicles after so many miles driven. Whenever a bus comes up on the threshold, mechanics will pull it from the fleet for maintenance.
Share23 May 2022
