Condition Monitoring is Expanding
As AI takes the wheel, warehouse equipment will experience less downtime.
Warehousing equipment is a big capital investment, one that usually pays off. But if you’re going to get the most of that investment—as well as an ROI—you want that equipment up and running as often as possible. Shutdowns and delays can be costly, not just the price tag for repairs but in lost business, too.
To combat that downtime in recent years, equipment manufacturers have added condition monitoring to their equipment. It works by collecting high quality data from sensors and making that data visible. For instance, if the temperature is rising on a piece of equipment, it will alert the maintenance team who can then determine the problem and fix it. While that’s a big advancement from the past, when equipment just stopped and you had to figure out what went wrong, there are even better options in the pipeline.
The next advancement in condition monitoring is coming in the form of predictive maintenance. Aided by AI and machine learning, this advancement takes the data from the drives directly and looks into the equipment to see what exactly is going on. If the prediction matches the condition, the operators can simply move on.
This varies from the current state of condition monitoring, which works something like your car and its maintenance calendar. Think about your car’s dashboard; when it’s time for scheduled maintenance it will alert you, and you take it in for whatever mileage servicing is due. However, there might not actually be anything that requires servicing. Without predictive analysis, you don’t know this. In the case of your warehousing equipment, you might regularly refill the oil every 100 miles, independent of the condition.
With predictive maintenance, you’ll instead be working off information that predicts failures or issues, allowing you to head them off at the pass.
Predictive maintenance capabilities are growing rapidly, although they are largely in a beta stage for now. OEMs are allowing machine learning to train on algorithms right now to detect system anomalies. The equipment is training on factors like motor current, temperature sensors, vibration, and anything related to moving parts. The hope is that eventually, motors, gear boxes and bearings will all be equipped with predictive maintenance capabilities coming directly out of the drive.
Once predictive maintenance capabilities are ready to roll out, customers will be able to take advantage of them and learn about their benefits. The likely scenario is that within five years, drives will be trained up and ready to deliver on predictive maintenance without additional sensors.
With a labor shortage, less equipment will mean less reliance on maintenance personnel, helping to alleviate the issue of manpower. Equipment costs will come down, too, as there will be less need for spare parts.
The power of computers is still growing, and each generation brings more to the table. For warehousing equipment the payoff will be less downtime, thanks to predictive maintenance.
For more information about the Solutions Community: mhi.org/solutionscommunity
For further articles from the Solutions Community:
Warehouse Execution Software Implementation
Mitigating Damage from Many Directions
Are You Data Rich, Intelligence Poor?
For a Successful Software Implementation, Focus on the People
