Which Operations Would Be Described as Preventive Maintenance?

There are several operations included in preventive maintenance. Mostly preventive maintenance is only associated directly with doing maintenance on and interacting with equipment assets, tools, motors, systems, and other critical parts. But in fact, there are several levels or operations attached to the phrase.

Preventive Maintenance Operations on Machines

Mapping of equipment assets

Mapping everything that needs maintenance is typically the first thing to do. Make a list of everything that will break apart given enough time. This will typically include almost all parts of the machine.

You can choose to limit the list to moving parts only, as it’s most likely only moving parts you’ll be doing preventive maintenance on. But if you want to do it right, you should be making a list of all parts including fixed parts. Fixed parts could be different nuts that are keeping vital parts working i.e.

Once you have the list you can start to prioritize them.

There are several different ways to prioritize. You can make a priority on how vital the part is for the machine to be working 100% or you could make a priority based on how often something should be changed. This will differ from machine to machine. Some machines will require all parts to be working for the machine to be working. Here will a simple priority based on change interval be fine. However, if your machine has parts where the machine could be functioning just fine for days or weeks, where the part is broken. You’ll need a priority on how vital the parts are to determine your maintenance plan.

A good approach to follow would be to make a priority list based on vital parts and then group the parts into maintenance intervals.

Let me give you a primitive example of what the grouping would look like:

Maintenance every 1 week

1. CNC Robot ARM
2. CNC Spindel/Rotor
3. Drill Bit
4. End Mill
5. Reamers
6. Hollow Mill

Maintenance every 4 weeks (monthly)

1. Reamers
2. Hollow Mill
3. Motorized/automatic door

Maintenance every 1 year

1. Conveyor Motor
2. Conveyor Belt
3. Robot Arm Stacker

Maintenance planning and systems

Once you’ve mapped all your assets that need scheduled maintenance you need to put them into a system. The system could in principle be a piece of paper, whiteboard, excel document, or software built for it.

Planning your maintenance can be categorized into 3 types of planned/preventive maintenance.

Time-based preventive maintenance
Usage-based preventive maintenance
Condition-based preventive maintenance

Time-based preventive maintenance

Planning maintenance based on a timed interval means that you’ll set up a specific date or time within a week, month or year. For example, time-based maintenance could be scheduled to be done every Monday at 17 pm, or on the 4th of August every year.

Time-based planning is great for things that must happen no matter the state of the machine. Cleaning or oil changes are good examples of this.

Usage-based preventive maintenance

Planning maintenance on a usage-based interval means that the scheduled maintenance will depend on how much the parts/assets have been used. Cars for example typically are on usage-based maintenance. The maintenance will happen once every 15.000 km. i.e. Here it’s extremely important to have tracking equipment that can monitor how long things have been running and based on this create alarms once reaching above a certain time.

Usage-based maintenance is harder to reach than time-based because it requires more than just a spreadsheet with dates. However, it is a more precise and effective approach to maintenance.

It’s in the usage-based category we see many SAS (software) products enabling you to track your maintenance and make alarms. It’s also in this category you’ll see a lot of machine learning getting introduced where we go more into usage-based predictive instead of preventive maintenance.

Condition-based preventive maintenance

I see two types of condition-based maintenance. One is, actually see how the condition of the equipment is in and base the maintenance plan on that. This usually requires a lot of experience with the equipment and can many times be faulty, because of not relying on data.

The second is having alarms that’ll indicate the condition itself. This can be done with many types of sensors. Typically you’ll see the vibration, ampere, and temperature data being used for this. When a motor is about to require maintenance it’ll typically vibrate more, require more power to run, and get hotter. Based on the data above you can make alarms that trigger when the measurements get above or below a certain threshold.

Here machine learning also gets used a lot, because over a long period when you have a lot of data for all scenarios, a machine will be able to learn what to look out for in the measurements. So for example, if it suddenly sees a rise in vibration together with a higher temperature, it’ll know if it’s normal or not.

Responsibility

Large cooperations typically have types of workers doing different maintenance to the machines. We typically see machine operators, electricians, mechanics, engineers, and managers doing different types of maintenance.

Operators usually do everyday tasks like cleaning, oiling, and changing small parts like tools within the machine. Electricians, mechanics, and engineers are doing maintenance on the machine itself. Managers typically have the responsibility for the planning of the maintenance.

Systems often forget the responsibility part. The different roles are placed in different environments and need the maintenance plan served in different manors.

Operators need direct alerts on the machine itself if anything needs to be done. Electricians, mechanics, and engineers typically need notifications through SMS or email, and maybe a scheduled plan in the calendar. Managers need the overview to be able to create, edit and delete scheduled maintenance, and on top of that be able to see who is doing what at what time. Here a simple spreadsheet or a piece of paper will be too primitive. To be able to handle all this, a good CMMS (Computerized maintenance management system) is needed.

Most CMMS are in reality just structured spreadsheets on drugs with many features that allow connection between other systems. CMMS is definitely needed if you are in a large cooperation with tens or hundreds of machines. However, getting simple alerts based on time, usage and conditions is relatively simple and cheap to achieve.

Reach out to hi@optipeople.dk to hear more about what we can do to reduce your maintenance costs.