Ignoring alert notifications is far too common in industry. During a visit to a grain mill we were asked if our system could interface to the plant monitoring system. Looking at the dashboard screen we noticed active alerts flashing and it was explained as "oh they have been there for a while. There's no real issue". Although that is surely true, what will happen when there is an issue with this component? It's also common to see alarm horns disconnected or unplugged and overloads bypassed. In a world of fighting fires and pressure on maintenance teams to restart production we sometimes cut corners fully intending to correct the issue later.
Setting of alert thresholds is a very important step that can take months to get right. It's mostly science but there can be an element of art to this as well. Some times there are variables that complicate this process including:
Variable speed applications - An example of this might be a cooling pump that automatically speeds up and slows down based on demand. It's important to use the highest speed conditions when setting alert thresholds to avoid nuisance alerts unless your system has speed input and a sophisticated algorithm that factors in speed.
Variable torque processes - Some machines are designed to process multiple products. Each product may put a different load on certain machine components. Again, the worst case condition should be used for alerts unless you have some way to measure or calculate torque such as current, magnetometer or an actual torque sensor.
Variability of temperature - For installations in non climate controlled environments, component temperature will vary seasonally. If the system is installed in the winter time and tight alert thresholds set based on these conditions, come summer you will experience alerts. You may have to adjust the thresholds to match summer conditions unless your system has a way to sense ambient temperature and automatically adjust threshold based on this.
One customer
No scheduled tasks -
If your day is to come in and only pay attention to the work orders, then the system might take a back seat. If the system takes a back seat you run a high risk of not catching critical notifications that something could be going wrong with your equipment. You may check it from time to time, but utilizing the monitoring system should always be a part of your scheduled tasks. Checking the signals and for sensor notifications should be a daily routine.
Training -
lack of engagement can stem from feeling intimidated by change or technology. Consistently working together to change how preventive maintenance is achieved involves preparation.
Comments