There are several ways in which this information can be applied. One way deals with instances where there are either one or a small number of assets making up the fleet. The best example of instances involving small numbers are plants, like process plants or refinery units.
Another way deals with instances where there are large numbers of assets making up the fleet. Examples of large number fleets are locomotives, vehicles, airplanes and other mobile assets.
Beginning with applications to small number fleets like plants, there are a number of characteristics that drive the way information is used and the relative importance of the information. In many situations involving plants or small number fleets, the owner of the asset is the developer and the custodian of the information. As a result, much of the information and its applications are developed over a longer period of time while the asset is in operation.
Identification of failure mechanisms and failure modes occur over time as the result of bad experiences. A wise and prudent owner will use those bad experiences as learning opportunities and not allow them to happen twice.
For instance, if the blades on a fin-fan fail due to fatigue, the owner will begin the response by determining the number of cycles when that failure occurred and then check all similar fans to see if they are nearing the same life. A prudent owner will go a step further and look at all other fin-fans in the plant to see if the blades are operating above their fatigue limit. If above the fatigue limit, the owner will estimate the number of fatigue cycles each fan has experienced. If nearing the end of life, the owner will take measures to prevent failure.
Extending the example described above to the development of predictive maintenance and preventive maintenance, a prudent owner will create a program of maintenance in which the number of fatigue cycles is closely monitored and fan blades are replaced at some point that allows a comfortable margin in time before fatigue cracks form and a failure occurs.
Since plants typically provide an opportunity for individuals to walk out to the site of failure to see the failure being reported, the system for reporting the failure has a different level of importance than those cases where the failure occurs at some remote site.
Large plants with lots of operating units and equipment typically have many opportunities competing for a moderate number of resources. Typically, the largest plant with the highest ranking and most influential staff will have first call on the resources unless an objective method of prioritization is applied to the distribution of scarce resources.
Small plants have fewer problems and fewer individuals competing for resources, but they also have far fewer resources. In small plants, it is equally necessary to have an objective system for prioritizing needs for the limited resources.
When developing new assets, the designers are frequently outside contractors who have little or no experience with real-life failures. Left to their own devices, they will design assets based on whatever they are being told by the companies who provide the equipment or components. Again, they have limited experience with failure mechanisms that actually exist in the plant, or the failure modes that result.
In designing new assets, it is important for the future owners to clearly identify the failure mechanisms they deal with every day and the failure modes that result. They also need to identify the amount of predictive maintenance and preventive maintenance that is acceptable and the reliability performance that is required.
Tip from Critical Connections: Linking Failure Modes and Failure Mechanisms to Predictive and Preventive Maintenance by Daniel Daley
