Failure is eminent in every organization, but it shouldn’t be. Failure comes in many forms, from bad hires, to underfunded budgets, etc.

Obviously, budgets have an impact on our programs, we shouldn’t deny that reality, and run-to-fail reactive programs keep budgets in an unpredictable state. Leaders find this challenging. Although emergency equipment breakage can happen, equipment failure can be costly as supply chains are in flux, even though we are recovering from the epidemic years.

Point in hand, circuit boards are still long lead items for many types of equipment.

Equipment maintenance and/or replacement has tell-tail signs to avoid the situation of run-to-failure embarrassment with leaders. How so?

Failure analysis is your solution!

This practice can change your program from a reactive program (run-to-failure) to a predictive (gold standard) program.

Here is a scenario, you have worked at a company for years and no money has been put into equipment and repairmen must be called out often just to put Band-Aids on the clunking machine. Is it really doing its job well? Probably not, but it is working (at least enough to stop people from complaining).

Until the unit fails again, and a technician is called again. And of course, failure always happens at the importune time. For instance, the sump pump only fails during heavy rain, then the basement floods. Or another instance, the HVAC only fails on the hottest days. These failures are also when the clients will complain the most.

There must be another way, right?

There is – perform a failure analysis.

An example of a real-life piece of equipment. It was installed in 1997, and the unit ran until almost 2018, see figure #1 on the next page. The red line indicates the expected manufacturers life.

When should this unit have been changed out or replaced?

The clear answer was before the yellow circled expenses. With the parts and the service calls the dollar cost for use has risen substantially.

Companies all around the world try to play the game of getting the most use out of a piece of equipment. This happens by making it last longer the expected. However, the depreciation has long been spent, in most cases the depreciation is over 5-7 years. And the warranty has been used up also.

Running a failure analysis can help make that decision based on math, not opinion.

In the image, figure #1, it appears that the unit was headed for a failure based on the number of calls in the year 2016 to 2017. At this point a decision needs to be made by the administrator.

Whether to replace or continue the dilemma.

Waiting for failure is not ideal. And often comes with a ton of questions from superiors. Don’t let that happen to you.

You can…

Step #1

You will need a little more data.

The data used for Fig. 1, were the number of work orders, the manufacturers expected life, and the costs for every work ticket, and the dates of every ticket. Figure #2 is the work order history for example.

This step requires a good CMMS (computerized maintenance management System). One that is actively used, and good data put in. As the saying goes “junk in, junk out”, so the date needs to be good data with a goal in mind.

Or it’s just data for data’s sake. There are several great systems out there depending on your industry. I’d be happy to share a few good ones.

Within this data entry be as detailed as you can because the next steps are where the cost factors and predicting come in.

To make the system work even better data will need to be collected on all the equipment you have to make sure averages are well established for the industry you are in.

Step #2

Find a techy person that is familiar with excel. Although most CMMS systems can do these calculations, a simple excel program that is standard on most computers is fine. Looking at figure #1 again, plot the dates and the number of tickets on a graph.

You should end up with something like the example. This is good. It’s horrible to see but an eye opener just at this step to understand what your maintenance department is doing. This simple graph alone can start conversations with your trades and line managers on making better decisions to avoid failure of equipment.

However, if you really want to take your program to the next level, finding the right point for your organization to replace units BEFORE failure expenses start move to step three.

Step #3

For this step you will need a batch of units. Run the same information for each then on the excel program look for the Bell Curve graph. Configure the data for that style and you should get the below graph, Figure #3. This example includes the piece of equipment we started with and 10 more in similar situations.

The bottom horizontal line is the number of years between end-of-life expectations and the beginning of critical failure.

For instance, in Figure #2 the dates used would be 2014 to 2016, before critical failure starts.

The graph indicates, at the peak of the Bell, that the organization gets roughly 2 years after the manufactures expected dates before failure starts to happen.

Step #4

Using this data You, or your frontline manager/staff have the knowledge to predict expenses for your operations budget several years out to replace an aging fleet of equipment.

Then your work really starts. Now you need to write and present a story to your budget holders on how your department wants to save the world and raise the bar. There should also be a follow up conversation with the trades to see if there is any outside factors that might contribute to early failure or in lengthening the time between expectancy and failure.

Some topics could include:

1. Water quality

2. Electrical surges

3. Air flow

4. High dust or debris

5. Amount of use

After all, a great service department should catch issues before they are issues.

Additional Resource of knowledge on this topic can be found in the books that were handy to me as I was writing this. See below: