Transcription
[Music] Hello, my name is Dave Buros. I'm a global product marketing manager for ABB Process Automation Service, and today we're going to learn about an important service concept called reliability-centered maintenance.
Now, why would we want to learn about reliability-centered maintenance? It's considered the world-class maintenance approach for process industries and manufacturing, and other industries as well. So, we're going to cover three main areas in this presentation. It's just an overview, but those three main areas are: equipment criticality and analysis, failure modes and effects analysis on the equipment, and some basics about how to implement reliability-centered maintenance.
So, let's start with the definition of reliability-centered maintenance. Reliability-centered maintenance is a high-level maintenance strategy that's meant to optimize all the assets in the facility. And it uses a variety of maintenance approaches to put together an overall approach that's meaningful for every different system or equipment category. And these maintenance strategies are optimized to really provide the most overall efficient functioning of the plant.
So, the notion of preventive maintenance is that the more preventive maintenance we do, as you can see here in this line, is curving up these bars that curve up here, then the fewer unplanned maintenance activities will be required. Those are breaking, fix, emergency repairs. What also happens, the more preventive maintenance that we do, the actual planned maintenance hours also come down. We're just doing things more effectively, and we see cost benefits across the board.
But what if we do too much preventive maintenance? Is that possible? It is possible. There's a sweet spot where we only want to do the right kind of maintenance at the right time. If we touch the equipment too much, we can actually induce failures that are costly, or we can just spend too much time maintaining equipment, and the actual cost will begin to creep up. So, we want to be right here in this sweet spot.
So, what did we have before reliability-centered maintenance? Well, first of all, we had reactive maintenance. Something breaks, and we go and fix it. The idea is to get that equipment back up as quickly as possible. There's no real preventive maintenance with that approach. Costs aren't really considered that carefully. It's popular because it's very easy to deploy. Something breaks, and we go fix it. It's very inefficient use of manpower and capital.
Well, we've already discussed preventive maintenance, of course. That's the next step. If we can address something that might break on a piece of equipment before it breaks, then we're a step ahead of the game. But as I also said, it's possible to do too much preventive maintenance. We don't want to do too much, but still, that's a step in the right direction. And it's usually time-directed. We do it on a regular basis. Every six weeks, we're going to lubricate a piece of machinery.
Then we go to proactive maintenance. This is the notion of looking at a production line or a piece of equipment even before it's put into operation and make sure that it's easy to access to maintain, or to look at equipment. Say, is that equipment, is that piece of equipment easy to maintain? Should we replace that equipment? Or if you're building a plant or building a new line, you can have that as a criteria of looking at new equipment. Is this equipment easy to maintain? That's proactive maintenance.
There's also predictive maintenance. This is when we use a variety of technologies such as infrared or thermography or lubrication analysis, different kinds of technology to assess a piece of equipment even while it's in operation. That might be able to determine certain things on that piece of equipment that we can't see with our naked eye or we can't hear that will alert us, "Hey, something needs to be addressed here."
Then finally, we get to reliability-centered maintenance. And as I mentioned before, reliability-centered maintenance takes advantage of all of those things and only chooses the right approach for the right piece of equipment. Even reactive maintenance or a strategy might be to not choose any maintenance approach at all. All right, strategy might be to just let a piece of equipment run to failure if it's low-cost and you have a hot spare. Maybe it's just too costly to try to touch that on a regular basis. Let it fail.
All right, so let's talk about the three main steps to effective reliability-centered maintenance. The first one is to make sure that you preserve equipment functionality by establishing criticality levels of the equipment. The second is to identify and prioritize failure modes. Once we understand what the equipment criticality is, then we can start looking at what's likely to fail on this piece of equipment and put together some plans to address it. And then thirdly, implement reliability-centered maintenance by choosing the correct approaches.
So, for criticality, let me just state this clearly. The first consideration for criticality should always be safety. Here we have an example of a refinery in Asia, I won't name the country, that exploded in 1992. And the explosion killed a number of people and injured many more. When the analysis was done on why the explosion occurred, it was determined that it was really a maintenance failure. You see there some of the reasons for why repeated ratcheting, reducing diameter of a gasket retainer. Well, I'm not that familiar with this case, but repeated ratcheting, that could be an example of too much preventive maintenance. So, the right maintenance approaches weren't taken here, and it led to a disaster.
So, as I said, safety is always number one when you're putting together criticality analysis. Environmental is up there in the same level of safety. The next consideration is production stoppage. If you're not making product, you're not making money. The third one is the maintenance expense. How costly is it to repair a certain piece of equipment? The fourth is the effect on the systems. Remember, systems is a collection of equipment. So, what is the effect if something breaks here? What kind of effect will a ripple effect have on other pieces of equipment? It could be a big effect. So, take that into consideration. And fifth, redundancy. Does this production or is this process have some redundant features so that if it fails in one area, it picks up in another area? That's lacking redundancy, then it could be a higher level of criticality.
We put those criticalities into certain levels. So, this is an example of criticality level number one. First of all, if a failure occurred here, then it could result in a loss of life or a high likelihood of resulting in loss of life or limb or permanent disability, high likelihood of an environmental emission, high likelihood of a very costly impact on production, and high likelihood of a costly equipment repair. Maybe there's no redundancy with that system. So, that would get a criticality level number one.
Then we just go down there. Criticality level number two might be a measurable impact on safety and environmental, or a less costly impact on production, or a less costly maintenance. You get the idea. Level three, just a potential, still something to watch out for, but not high likelihood. And then number four, no safety effects, no environmental effects, no impact on production, very minimal, minimal maintenance costs, and maybe there's multiple redundancy in this one. So, you just don't have to worry that much about it.
So, those are criticality levels. And before we can assign criticality, we need to establish the systems. Systems is a group of components. And an equipment category, a piece of equipment, and all the subcomponents, or it could just be one singular complex piece of equipment. Here's an example. We'd have a major system with major sub-assemblies, major sub-assemblies, components, and we get down to a part level. For example, a compressed air system. We'd have compressor assemblies, compressor units. Each compressor unit would have a motor, a compressor, and controls. And then you get down to a parts level with a motor. So, we want to identify all the major components of a system, group them together if we can, just in a sensible fashion.
So, for instance, we would have with that same kind of example, we would have in our failure modes and effects analysis worksheet, we'd have the compressed air system, looking at the subsystem of air compressor assembly. And the components would be compressor, motor, and valve. Then we want to identify the failure modes. What could go wrong with each of those components? So, that for a compressor, for instance, it might seize. And that would be a failure. For a motor, might fail to start. Failure. Valve fails to close. Failure.
Next, we want to look at the failure mechanisms. What would cause these things to fail? Here's just some examples. Could be corrosion or erosion, or a part fails, or it's not calibrated. So forth. So, again, in our example with a compressor, the failure mode is that it seizes. And what would cause that? Might be a bearing failure. So, we just go through all of our systems like that and fill out this sheet. And we create a likelihood table of that happening with a level such as this: remote, very low, low, high, up to extreme.
We also want and effects descriptions. So, if we think about the levels that we defined earlier when we were talking about levels of criticality, level one criticality would be loss of life, body part, or lost-time accident, and so forth. So, we want to identify those specifically because we want to apply, we want to have an apples-to-apples comparison with all the systems in the plant. So, again, back to our example, we have with a compressor, we have the likelihood of it happening as ranked high. And we also have the failure effect also ranked high. It would have a major impact on production.
Then we want to put together the actions that we would take if there were a failure. So, we put together a list of recommendations. If a compressor ceases as a result of a bearing failure, we know that that's high likelihood and also high negative impact on production. So, the recommended action is to maintain lubrication.
So, that was just a very cursory overview of what criticality analysis and failure mode and effect analysis is. Now, we'll talk briefly about how a plant would actually implement reliability-centered maintenance. First of all, you want to make sure you have a team, a cross-functional team. You would have that include mechanical maintenance, instrumentation, electrical maintenance, process control, operations, systems, our plant engineering, and perhaps a facilitator.
You want to understand your current maintenance processes. After you've done that criticality analysis and the failure mode and effect analysis, you might have identified areas where your processes just wouldn't allow you to really effectively act on the recommendations that you just put together as part of the failure modes and effects analysis. You want to put together a plan then to address those areas that might be weaknesses in your processes.
Start by setting goals. What we really want to do here? Do we want to increase production? What are the plant's business goals? Are we starting up a new line and so we want to get it right the first time? Do we need to reduce maintenance costs? Put together your approach. How are you going to track it? How are you going to, who's going to be responsible for what? Make sure that people know that they're accountable. So, there's a name for certain actions so it doesn't fall into the cracks.
Finally, implement into an enterprise asset management system, also known as a computerized maintenance management system. This is software where work orders can be implemented, and the preventive maintenance routines can be in there, and the planning and scheduling can be done there, and all the things that we've just discussed can all be implemented in there and expedite effective implementation of reliability-centered maintenance.
Now, ABB has its own computerized maintenance management system for ABB equipment, and that's called Service Pro. But Service Pro is uniquely designed for ABB use, for ABB equipment and processes, and it comes already with a hierarchy of systems and subsystems and components that you see here. It also comes with the preventive maintenance routines already embedded in there. So, there's things that we saw that are a result of a failure modes and effects analysis. Many of those things are already embedded in the Service Pro for many ABB asset classes.
Service Pro is also very helpful for customers in doing their criticality analysis. What you see here is an analysis of the service events and the number of hours spent maintaining specific pieces of equipment. You see here that Analyzer 1200 has had a high number of service instances and also a high amount of service hours. Well, that'll tell the customer, perhaps that's a critical piece of equipment that we need to put in a different level, a higher level of criticality, or maybe that's a piece of equipment that needs to be addressed or redesigned or even replaced.
Service Pro also provides benchmarking for specific pieces of equipment at other sites around the world that had that same kind of, that same piece of equipment. Since Service Pro is used by ABB engineers and customers at hundreds of sites all over the world, a customer can look at this kind of data and see the piece of equipment, my plant needs more repair than that same piece of equipment at many other customer sites. Is there something wrong with the way I'm maintaining it? Does that piece of equipment need to be replaced? All of this is an important contribution to a criticality analysis and a failure modes and effects analysis.
Another tool that ABB has that will help customers very much in doing a criticality analysis and a failure modes and effects analysis as part of a reliability-centered maintenance implementation is Advanced Services powered by Service Port. As many of you know, when we deliver advanced services, we are gathering lots of data and analyzing that data and identifying, categorizing, and prioritizing things that need to be addressed. The same thing that happens in a failure modes and effects analysis. Here is a typical screen that you'd see in a fingerprint report or on the Service Port itself that would show the systems or subsystems or components we are monitoring or gathering data from with the advanced service. And with this size bar, we see events that might indicate that that piece of equipment needs to be looked at more closely, or that component needs to be looked at more closely. So, again, that would be a great contributor to a customer undertaking a reliability-centered maintenance program. We'd love to be able to provide that, that kind of analysis for customers for all of their systems. Right now, we only offer them for certain important ABB asset classes. We provide that kind of analysis via Advanced Services powered by Service Port for the Harmony control system, System 800xA, LV drives, mine hoists, and quality control systems for the paper industry.
So, that's just a brief overview of reliability-centered maintenance. Thank you for your time.
[Music]
[Music]